Final Report on Electric Clothes Dryers and Lint Ignition Characteristics May 2003

Transcription

1 Final Report on Electric Clothes Dryers and Lint Ignition Characteristics Abstract In FY 22, U.S. Consumer Product Safety Commission staff completed a test program to evaluate residential electric clothes dryers under various test conditions. The test program included measurements of temperature and airflow characteristics for different electric clothes dryer designs under normal operating conditions and conditions of partially-blocked and fully-blocked exhaust ducting. The ignition characteristics of lint in relation to electric clothes dryer operation were also evaluated. The test program produced data on different dryer operating and design characteristics that can be used to help reduce the risk of lint ignition in a clothes dryer and help prevent fires. The experiments described in this research report were undertaken to support future advances in clothes dryer safety. This report should not be used to suggest that current clothes dryers are unsafe or defective.

2 IMPORTANT All airflow measurements are reported in Standard Feet per Minute (SFPM). These measurements are referenced to 25ºC and inches Hg.

3 FINAL REPORT ON ELECTRIC CLOTHES DRYERS AND LINT IGNITION CHARACTERISTICS The experiments described in this research report were undertaken to support future advances in clothes dryer safety. This report should not be used to suggest that current clothes dryers are unsafe or defective. U.S. CONSUMER PRODUCT SAFETY COMMISSION 433 EAST WEST HIGHWAY, BETHESDA, MARYLAND 2814

4 U.S. CONSUMER PRODUCT SAFETY COMMISSION DIRECTORATE FOR ENGINEERING SCIENCES FINAL REPORT ON ELECTRIC CLOTHES DRYERS AND LINT IGNITION CHARACTERISTICS Arthur Lee Electrical Engineer Division of Electrical Engineering Directorate for E S ngineering ciences i

5 ACKNOWLEDGMENTS This project was completed with the help of CPSC staff in the Directorate for Laboratory Sciences; Directorate for Engineering Sciences; Directorate for Economics Analysis; and Directorate for Epidemiology. Frank Dunmore, Ph.D., Directorate for Laboratory Sciences, provided technical expertise in all aspects of the project, and specifically in coordinating and implementing Task 3 - Monitor Lint Distribution. He also authored the section of this report that discusses Task 3. Timothy Smith, Directorate for Engineering Sciences - Division of Human Factors, conducted an analysis of consumer behavior with regard to clothes dryers, particularly related to abnormal operating conditions. Sheela Kadambi, Directorate for Engineering Sciences, Division of Electrical Engineering, conducted the initial work on clothes dryers in 1998, which set the foundation for this project. Terrance R. Karels, Directorate for Economic Analysis, provided information on the clothes dryer market over the past 1 years. Mark Levenson, Ph.D., Directorate for Epidemiology Division of Hazard Analysis, developed estimates of clothes dryer fire losses for a 1-year period and analyzed in-depth investigations of fire incidents associated with clothes dryers. ii

6 EXECUTIVE SUMMARY The U.S. Consumer Product Safety Commission (CPSC) initiated a project in Fiscal Year 22 to investigate possible conditions that may lead to dryer fires and to develop recommendations for revisions and/or additions to the voluntary standards to address potential hazards. In 1998, CPSC estimates that there were approximately 15,6 clothes dryer fires resulting in 2 deaths, 37 injuries and $75.4 million in property damage 1. CPSC staff tested clothes dryers to evaluate the effects of lint accumulation and abovenormal operating temperatures and determine whether such conditions may result in lint ignition and/or dryer fires. The data was used to help determine if dryer fires result from a single event or a combination of events. The basic approach was to conduct several tasks that could link the cause of lint accumulation to possible dryer fires and/or lint ignition. The tasks included: Task 1. Inspect and Record Dryer Design Task 2. Document Dryer Operating Characteristics Task 3. Monitor Lint Distribution Task 4. Determine Characteristics Required for Lint Ignition Although selected dryer designs were used to document the variety of temperature and airflow patterns in a dryer, the conclusions are based on and can be applied to general dryer designs. Tests included examining the effects of restricted and unrestricted airflow on dryer operation. Airflow restriction was created by placing an iris which created blockages of 25, 5 and 75 percent in the exhaust duct. A blast plate covering the exhaust opening was used to create a fully blocked exhaust vent (1 percent blockage). Hot wire anemometers were used to measure the airflow entering and exiting the dryers. Thermocouples were placed at the heater intake, heater housing, heater exhaust, tumbler intake, blower intake, and dryer exhaust. The results of the CPSC staff tests showed that lint that accumulates inside the dryer can ignite if the lint contacts certain areas of the heater housing, if the lint is in proximity to the heater, or if the lint is ingested by the heater box. Observations made for each task during testing include: Task 1. Inspect and Record Dryer Design All four dryer designs used the same method and order (heater, tumbler, lint screen, blower, and exhaust duct) for moving the air through the dryer. 1 Mah, J., Table 1, Estimated Residential Structure Fires Selected Equipment 1998, 1998 Residential Fire Loss Estimates, Directorate for Epidemiology, US Consumer Product Safety Commission, iii

7 The length of the dryer s exhaust duct extending out of the dryer may not allow the house duct to slide far enough onto the dryer s exhaust duct to provide a secure pressure fit. Using rigid external ducting does not allow for a secure pressure fit around the dryer s male duct. Task 2. Document Dryer Operating Characteristics The temperatures within a dryer, under both normal and abnormal conditions, were similar for the four different dryer designs tested, with only slight variations due to dryer internal configurations. The temperatures measured inside the heater box, heater intake, and intake into the tumbler increased when the exhaust vent was partially blocked or fully blocked. The temperatures inside the tumbler, blower and exhaust vent decreased when the exhaust vent was partially blocked or fully blocked. When the exhaust vent was blocked up to 5 percent, the temperatures inside the dryer were similar to those measured when there was no blockage of the exhaust vent. When the exhaust vent was 75 percent or 1 percent blocked, temperatures in certain areas inside the dryer increased significantly. Under normal operation, the airflow inside the exhaust vent decreased dramatically as the lint screen became blocked with lint particles. In general, the dryers only cycled on the high-limit thermostat when the exhaust vent was 75 or 1 percent blocked, which caused the temperatures near the heater to increase significantly. When the primary thermostat was bypassed (simulating a thermostat failure), the dryer operated at higher than normal temperatures temperatures similar to those measured when the exhaust vent was blocked 5 to 75 percent. In general (3 of the 4 dryer designs tested), a failed-closed primary thermostat did not cause the dryer to cycle on the high limit thermostat for the unblocked exhaust vent condition. Task 3. Monitor Lint Distribution Lint begins to accumulate inside a dryer chassis upon first use. Lint accumulates on the dryer s components, including the heater and the dryer floor. This accumulation occurs even when the dryer s lint screen has been cleaned after each usage, and the dryer is properly exhausted. Seals in the dryer s interior exhaust venting may not be adequate to prevent linty air from escaping into the dryer s interior. iv

8 Task 4. Determine Characteristics Required for Lint Ignition Lint that accumulates on the heater housing can easily ignite under conditions of a failed high-limit thermostat and a blocked exhaust vent. Lint accumulating near the heater intake can ignite before the high-limit thermostat switches the heater element off. Lint ingested by the heater and embers expelled from the heater outlet can easily ignite additional lint or fabric in the air stream, resulting in additional embers in the dryer system and exhaust vent. The CPSC staff noted the following during testing and analysis: The high-limit thermostat may prematurely fail when subjected to high ambient temperatures. v

9 TABLE OF CONTENTS ACKNOWLEDGMENTS...ii EXECUTIVE SUMMARY...iii 1. INTRODUCTION General Project Objective Focus Objective Technical Approach Task 1: Inspect and Record Dryer Design Task 2: Document Dryer Operating Characteristics Task 3: Monitor Lint Distribution Task 4: Determine Characteristics Required for Lint Ignition Ignition of Lint on and near the Heater Housing Ignition of Lint Ingested into the Heater Organization of the Report Statement of Test Methodology Global Terminology DESCRIPTION OF TESTS AND TEST RESULTS Task 1: Inspect and Record Dryer Design Airflow Pattern Heater Location and Configuration Lint Screen Location Tumbler Design Blower (Fan) Design Operating Features Safety Device Locations Task 2: Document Dryer Operating Characteristics Exhaust Vent Setup Instrumentation Setup Global Test Procedure and Setup Dryer Design A Dryer Design A Temperature and Airflow Characteristics, Blocked and Unblocked Exhaust Vent Conditions Dryer Design A Airflow Dryer Design A Primary Thermostat Bypassed Dryer Design B Dryer Design B Temperature and Airflow Characteristics, Blocked and Unblocked Dryer Design B Airflow Dryer Design B Primary Thermostat Bypassed Dryer Design C Dryer Design C Temperature and Airflow Characteristics, Blocked and Unblocked Dryer Design C Airflow Dryer Design C Primary Thermostat Bypassed Dryer Design D Dryer Design D Temperature and Airflow Characteristics, Blocked and Unblocked vi

10 Dryer Design D Airflow Dryer Design D Primary Thermostat Bypassed Task 3: Monitor Lint Distribution Test Setup Test Description Method 1 Weight Loss of Load Method 2 Visual Examination Examination of Dryer Design A Task 4: Determine Characteristics Required for Lint Ignition Test Setup Instrumentation Setup Phase I Ignition Characteristics of Lint On and Near the Heater Box Top of Heater Housing Side of the Heater Housing Lint Samples at the Heater Intake Phase II Ignition Characteristics of Lint Ingested into the Heater Box Part 1 Lint Samples Ingested by the Heater Box Part 2 Ignition of Target Materials Downstream from the Heater Exhaust DISCUSSION Task 1: Inspect and Record Dryer Design Task 2: Document Dryer Operating Characteristics Normal Operation (Unblocked Exhaust Vent) Partially-Blocked and 1%-Blocked Conditions Task 3: Monitor Lint Distribution Dryer Design A at Positive Pressure Other Dryer Designs at Positive Pressure Task 4: Determine Characteristics Required for Lint Ignition Lint on the Heater Housing Lint Samples at the Heater Intake High-Limit Thermostat Bypassed and No Airflow High-Limit Connected in Series and No Airflow High-Limit Connected in Series and Airflow Lint Samples Ingested into the Heater Ignition of Target Material Downstream from the Heater Exhaust High-Limit Thermostat Analysis High-Limit Thermostat 1 (HLT1) Analysis High-Limit Thermostat 2 (HLT2) Analysis SUMMARY AND CONCLUSIONS Appendices A to J...A-J vii

11 LIST OF FIGURES Figure 1. Dryer Design Characteristics...6 Figure 2. Heating Element in a Heater Box Configuration...7 Figure 3. Circular Heating Element in Dryer Design D...7 Figure 4. Exhaust Vent Setup...9 Figure 5. Thermocouple Response Before Task 2 Testing...11 Figure 6. Thermocouple Response After Task 2 Testing...11 Figure 7. Average Responses per Oven Set Temperatures and Day...12 Figure 8. Dryer Design A Intake and Exhaust...14 Figure 9. Dryer Design A Thermocouple and Airflow Measurements with a Wet Load...16 Figure 1. Dryer Design A T1 Heater Intake Comparison...17 Figure 11. Dryer Design A T3 Heater Housing Comparison...18 Figure 12. Dryer Design A T2 Heater Exhaust Comparison...19 Figure 13. Dryer Design A Intake Air Velocity...21 Figure 14. Dryer Design A Exhaust Air Velocity...22 Figure 15. Dryer Design A Average Airflow Velocities (High Heat and Dry Load)...23 Figure 16. Dryer Design A Average Airflow Velocities (No Heat and No Load)...24 Figure 17. Dryer Design A Primary Thermostat Bypassed...26 Figure 18. Dryer Design B Air Flow Through the Tumbler...27 Figure 19. Dryer Design B Thermocouple and Airflow Measurements (Wet Load)...3 Figure 2. Dryer Design B T1 Heater Intake Comparison...31 Figure 21. Dryer Design B T3 Heater Housing Comparison...32 Figure 22. Dryer Design B T5 Intake into the Blower Comparison...33 Figure 23. Dryer Design B Intake Air Velocity...35 Figure 24. Dryer Design B Exhaust Vent Air Velocity...36 Figure 25. Dryer Design B Average Airflow Velocities (High Heat and Wet Load)...37 Figure 26. Dryer Design B Primary Thermostat Bypassed...39 Figure 27. Dryer Design C Airflow through the Tumbler...4 Figure 28. Dryer Design C Thermocouple and Airflow Measurements (Wet Load)...43 Figure 29. Dryer Design C T1 Heater Intake Comparison...44 Figure 3. Dryer Design C T3 Heater Housing Comparison...45 Figure 31. Dryer Design C T5 Intake into the Blower Comparison...46 Figure 32. Dryer Design C Intake Air Velocity...48 Figure 33. Dryer Design C Exhaust Vent Air Velocity...49 Figure 34. Dryer Design C Average Airflow Velocities (High Heat and Wet Load)...5 Figure 35. Dryer Design C Primary Thermostat Bypassed...52 Figure 36. Dryer Design D Airflow Through the Tumbler...53 Figure 37. Setup with a Relay and Battery for T2 Analog...54 Figure 38. Dryer Design D Thermocouple and Airflow Measurements (Wet Load)...56 Figure 39. Thermocouple, Airflow and Relay (T2 Analog) Measurements...57 Figure 4. Dryer Design D T1 Heater Intake Comparison...58 Figure 41. Dryer Design D T3 Heater Housing Temperature Comparison...59 Figure 42. Dryer Design D Intake Air Velocity...61 Figure 43. Dryer Design D Exhaust Vent Air Velocity...62 Figure 44. Dryer Design D Average Airflow Velocities (High Heat and No Load)...63 Figure 45. Dryer Design D Primary Thermostat Bypassed...65 Figure 46. Dryer Design A Used in Task 3 Testing...66 viii

12 Figure 47. Lint Accumulated at Bottom of Dryer After 1 cycles...68 Figure 48. Lint Accumulated on Interior of Dryer After 48 Cycles...69 Figure 49. Lint Accumulated on Interior of Dryer After 1 Cycles...69 Figure 5. Lint Build-up at Blower Housing Seal and Dryer Floor...7 Figure 51. Test Setup for Lint Ignition...73 Figure 52. Test Setup for Lint Ignition...74 Figure 53. Location of Lint Samples on the Housing...75 Figure 54. Lint Sample on Heater Housing at Location A...76 Figure 55. Sample Smoldering at Location A...77 Figure 56. Sample s5-8 at Location B...78 Figure 57. Thermocouple Data for Sample s Figure 58. Sample s2-7 on the Side of the Heater Housing at Location C...8 Figure 59. Sample s2-7 Thermocouple Traces...8 Figure 6. Sample s2-2 Tested at Location D...81 Figure 61. Sample s2-6 Tested at Location E...82 Figure 62. Thermocouple Traces for Sample s2-6 at Location E...83 Figure 63. Lint Sample at Heater Intake...84 Figure 64. Infrared Camera Setup...84 Figure 65. Sample s3-1, Video Camera and Infrared Camera...87 Figure 66. Sample s3-1 Thermocouple and Airflow Data...88 Figure 67. Setup with Glass Tube...89 Figure 68. Setup with Target Material in the Glass Tube...91 Figure 69. Target Sample Igniting During Test Figure 7. Test 1 Thermocouple and Airflow Measurements Task 4- Test Figure 71. Dryer Design Comparison for T1, Heater Intake...97 Figure 72. Dryer Design Comparison for T3, Heater Housing...97 Figure 73. Dryer Design Comparison for T5, Blower Intake...98 Figure 74. Dryer Design Comparison for T4, Exhaust Vent...98 Figure 75. Maximum Temperatures Measured at Location...99 Figure 76. Intake into the Heater...11 Figure 77. Heater Exhaust Temperature...11 Figure 78. Heater Housing Temperature...12 Figure 79. Exhaust Vent Temperature...12 Figure 8. Intake into the Blower Temperature...13 Figure 81. Intake into the Tumbler Temperature...13 Figure 82. Maximum Temperatures for All Dryers at Each Location and Condition...15 Figure 83. Comparison of Exhaust Air Velocity for all Dryer Designs...17 Figure 84. Calculated Average Exhaust Air Velocity...17 Figures 85 (a) and (b). Dryer Design A Blower Assembly...18 Figure 86. An Illustration of Airflow through the Dryer and Blower...19 Figure 87. Junction Between the Blower and Exhaust Vent...11 Figure 88. Lint Samples Tested 3 Inches from the Heater Housing Edge Figure 89. T1 Only - 3 Inches from the Heater Housing Edge Figure 9. T2 Only - 3 Inches from the Heater Housing Edge Figure 91. T3 Only - 3 Inches from the Heater Housing Edge Figure 92. Lint Samples Tested 4 Inches from the Heater Housing Edge Figure 93. Ignition Time for 2 and 3 Inches from Heater Edge Figure 94. Lint Samples s3-16 and s3-17 Tested 1 Inch from the Heater Edge Figure 95. Lint Samples s3-14 and s3-18 Tested 2 Inches from the Heater Edge Figure 96. High Limit Activation Time Comparison at Sample 2 and 3 inches ix

13 Figure 97. Lint Samples Tested 1 Inch from the Heater Edge with Airflow...12 Figure 98. Samples Tested at 1 Inch, With and Without High-Limit Thermostat Figure 99. Sample s3-13 Test with the High Limit Thermostat in Series Figure 1. Failed Thermostat compared to a Replacement Thermostat Figure 11. X-ray Images of High-Limit Thermostat 1 (HLT1) Figure 12. Contacts for High-Limit Thermostat 1 (HLT1) Figure 13. Contact Pads for High-Limit Thermostat 1 (HLT1) Figure 14. Sample s5-14 with High Limit Thermostat in Series Figure 15. High-Limit Thermostat 2 (HLT2) before and after Failure Figure 16. X-ray Images of High-Limit Thermostat 2 (HLT2) Figure 17. Internal View for High-Limit Thermostat Figure 18. Setup for Testing High-Limit Thermostats...13 Figure 19. New High Limit Thermostat...13 Figure 11. Thermocouple Data for the New High-Limit Thermostat Figure 111. Second Failed High Limit Thermostat, HLT Figure 112. Comparison between New and Failed High Limit Thermostats LIST OF TABLES Table 1. Unrestricted and Restricted Duct Configurations...9 Table 2. Equivalent Linear Duct Length...1 Table 3. General Locations of Thermocouples and Anemometers...12 Table 3-continued. General Locations of Thermocouples and Anemometers...13 Table 4. Dryer Design A Measurement Data ( C and sfpm)...15 Table 5. Dryer Design B Statistics ( C and sfpm)...28 Table 6. Dryer Design C Statistics ( C and sfpm)...41 Table 6. Thermocouple Location and Setup...71 Table 7. Testing on Top of the Heater Housing at Location A...76 Table 8. Testing on Top of the Heater Housing at Location B Table 9. Testing on the Side of the Heater Housing at Location C...79 Table 1. Testing on Top of the Heater Housing at Location D...81 Table 11. Testing on the Side of the Heater Housing at Location E...82 Table 12. Lint Samples at the Heater Intake...86 Table 13. Lint Samples for Task 4 - Phase II Testing...9 Table 14. Target Materials for Task 4 - Phase II Testing...91 Table 15. Change in Minimum Temperature...16 Table 16. Change in Average Temperature...16 Table 17. Change in Maximum Temperature...16 Table 18. Ignition Times at Heater Intake Table 19. Approximate High Limit Thermostat Activation Times Table 2. Life History of Failed High-Limit Thermostats x

14 1. INTRODUCTION 1.1 General Over the years, the number of safety devices included in clothes dryers to reduce the incidence of fire has increased. The increased safety of dryers has likely contributed to the slowed growth rate in dryer fires. However, there were an estimated 15,6 clothes dryer fires resulting in 2 deaths, 37 injuries and $75.4 million in property damage in In 1998, 56.2% of all U.S. households had an electric clothes dryer, and 17.9% had a gas dryer 3. In total, over 74% of all U.S. households had at least one clothes dryer in There are just over 1 million US households; thus, there are over 74 million clothes dryers in use in the U.S. Over the period 199 through 1998, shipments (gas, electric, and compact dryers) increased 41% overall, from almost 4.6 million units to almost 6.5 million units 3. In 21, the number of dryer shipments was over 6.7 million units 4. There was a consistent growth in the sales of clothes dryers in the U.S. from 199 to 21. The U.S. Consumer Product Safety Commission (CPSC) initiated a project in Fiscal Year (FY) 22 to investigate possible conditions that may lead to dryer fires and to assess the adequacy of the applicable voluntary standards in addressing potential hazards. Four years earlier, in FY 1998, CPSC staff began an initial evaluation of clothes dryers. The results of the tests conducted during that evaluation showed that, when the dryer exhaust was blocked, some areas of the dryer would run cooler than normal and other areas would run hotter. CPSC staff believed that long-term operation of a dryer under conditions of restricted airflow, such as that caused by lint accumulation, could eventually lead to premature failure of components that may result in a fire. 1.2 Project Objective The CPSC clothes dryer project was initiated to determine the cause(s) for clothes dryer fires and to develop recommendations for revisions and/or additions to the voluntary standards to address those causes and help prevent dryer fires. 1.3 Focus Objective The focus objective of this project was to evaluate the effects of lint accumulation and above-normal operating temperatures in electric clothes dryers and determine whether such conditions may result in dryer fires and/or lint ignition. 2 Mah, J., Table 1, Estimated Residential Structure Fires Selected Equipment 1998, 1998 Residential Fire Loss Estimates, Directorate for Epidemiology, US Consumer Product Safety Commission, Appliance Magazine, Statistically Review, A Dana Chase Publication, May Appliance Magazine, U.S. Shipment Statistics, A Dana Chase Publication, March 22 1

15 1.4 Technical Approach The overall goal was to determine whether lint accumulation could result in clothes dryer fires and/or lint ignition. To accomplish this large task, smaller tasks were designed to link the cause of lint accumulation to possible dryer fires and/or lint ignition. The tasks were set up to eliminate as many dryer design variables as possible. Tests were conducted with sample clothes dryers at both normal and above-normal operating temperatures; with dryers operating in the high-limit cycle mode (caused by either a fully or partially blocked exhaust vent); with bypassed temperature limiting devices (simulating component failure); and without any safety temperature limiting devices. The following four smaller tasks were defined to accomplish the main objective: Task 1: Inspect and Record Dryer Design The objective of this task was to record the types and locations of the internal components and layout for each sample dryer and to photograph the internal configuration of each dryer. The airflow path throughout each dryer was illustrated and the locations of the heater, blower, and lint screen were documented. Based on the results of the examination of the dryers, the location of instrumentation was determined for Task 2 - Document Dryer Operating Characteristics Task 2: Document Dryer Operating Characteristics The objective of this task was to characterize the airflow patterns and temperatures inside each dryer design during normal (unblocked) and restricted (blocked) airflow through the exhaust vent. The dryers were characterized by operating according to the manufacturer s instructions with a (wet and/or dry) clothes load and without a load (a dryer may have been fully characterized with a wet load if the test data developed for a dry load and wet load, with no exhaust blockage, were not comparable.) The data in this task was used to set the test variables in Task 4 - Determine the Characteristics for Lint Ignition Task 3: Monitor Lint Distribution The objective of this task was to monitor lint distribution and accumulation in areas within a clothes dryer during operation with normal airflow. The load for these tests consisted of 1 wet (washed and spun dry) bath towels. A dryer was operated according to the manufacturer s instructions and subjected to normal use drying cycles. The dryer was operated for a total of 1 cycles. After 48 cycles, the accumulation of lint was recorded near the heater housing, internal exhaust duct, and internal dryer floor. The dryer was examined at the end of 1 cycles. Lint accumulation was characterized in one dryer. An analysis of potential causes for lint accumulation in other dryer designs was conducted. 2

16 1.4.4 Task 4: Determine Characteristics Required for Lint Ignition This task had two objectives: Evaluate the ignition characteristics of lint samples on and near the heater box, and evaluate the ignition characteristics of lint when samples were ingested into the heater box. The lint samples used in these tests were 1% cotton and were taken from the lint that accumulated during tests conducted in Task 3, unless otherwise noted Ignition of Lint on and near the Heater Housing The heat output and airflow through a heater were varied. Lint samples were placed at various locations on and near the heater box. Power to the heating element was introduced instantaneously or was stepped up to observe the different effects on the lint samples. Some tests were conducted with the high limit thermostat in series with the heating element, and some were conducted with the thermostat bypassed. The test results were categorized as one of the following: no ignition, charred only, smoldered, or ignition (flames) Ignition of Lint Ingested into the Heater Lint was introduced into the airflow of the heater intake. The lint samples were placed in front of the heater opening and restrained until ready for release. The system (heater, temperature, and air velocity) was stabilized for 5 minutes before proceeding. When the system had stabilized, the lint samples were released. The results were observed for 15 minutes, or less if the samples were consumed. 1.5 Organization of the Report This report is presented in two parts. The first part discusses the overall testing program and includes pertinent test descriptions and resultant data, analysis, findings and conclusions. The organization is such that each task is headed as a major test phase. The second part contains appendices that present expanded test data to support the findings and the conclusions. The appendices are also contained on the compact disk (CD), which can be accessed through a sub menu. 1.6 Statement of Test Methodology The test program was designed to identify and eliminate as many dryer design variables as possible for Task 4 testing. During the tests, temperature and airflow characteristics of each dryer design were recorded. Observations regarding dryer designs that may have caused variances in the test data are noted in the report. A large amount of test data was collected during this test program. Only the pertinent data for each task are presented. All data collected are noted in the report but may not be presented if the data are not pertinent to the discussion or conclusions. The clothes dryers used in this test program were selected by design, cost and features available. Different dryer designs were selected to demonstrate the variety of temperature and 3

17 airflow patterns in a dryer. The dryers selected were in the price range of $3 to $4 and included similar selectable dryer settings and features. Although only electric dryers were tested in this program, many of the conclusions may be applicable to gas-fueled clothes dryers. 1.7 Global Terminology Lint Units Top Front Rear Floor Inside Smolder Ignition Test Load Wet Load Dry Load Damp Load 1% cotton fibers that were expelled from a clothes load of cotton terry towels during the drying process. The lint material may refer to the material collected from the lint screen or the material that accumulated inside the cabinet (housing) of the dryer. Unless otherwise specified, all temperatures reported are in degrees Celsius ( C), all airflow measurements are in standard feet per minute (sfpm), and all weight measurements are in grams (g), and length in inches (in) and feet (ft). The side of the dryer that is viewed from above. The side of the dryer containing the door. The side of the dryer containing the dryer s exhaust vent and power cord. The internal side of the bottom of the dryer that rests on the building floor. The interior of the dryer containing the motor, drum, blower, and heater. May contain one or more of the following: smoke, embers, or charring with no evidence of flames. Flames are visible. The load of towels used in the dryer, either dry or wet load. Ten 1% cotton terry towels that were washed and spun dry in a washing machine, unless otherwise specified. No detergent was used. Ten 1% cotton terry towels that were dry to the touch, unless otherwise specified. Ten 1% cotton terry towels that were wet to the touch, but weighed substantially less then a wet load, unless otherwise specified. No detergent was used. High Limit Thermostat The thermostat located near the heating element of the dryer (High-Limit Switch). 4

18 Primary Thermostat The thermostat located between the lint screen and the blower (Operating Switch, Operating Thermostat). Thermostat Opened The thermostat reached its upper set point temperature and separated the contacts. Thermostat Closed The thermostat reached its lower set point temperature and closed the contacts. Abnormal Operating Condition The dryer operating or cycling on the high-limit thermostat, or tests conducted with all devices bypassed. Normal Operating Condition The dryer operating or cycling on the primary thermostat. High-Limit Cycling The dryer operating or cycling on the High-Limit Thermostat. Exhaust Vent The venting from the dryer to the outside of the building. All venting material was 4-inch rigid metal duct. All joints were sealed with foil tape except for the connection to the dryer; the exhaust vent was secured to the dryer using a 4-inch duct (hose) clamp. Fully Blocked Exhaust The exhaust vent was completely obstructed with a blast plate (1% blocked). Partially Blocked Exhaust The airflow in the exhaust vent was restricted by an iris opening less than 4 inches in diameter (25%, 5%, and 75% blocked). Unblocked Exhaust The exhaust vent contained no obstruction. Heater Heater Box The mechanism to warm the air flowing through the dryer (the heating element). A rectangular shaped housing containing the heating element. 5

19 2. DESCRIPTION OF TESTS AND TEST RESULTS 2.1 Task 1: Inspect and Record Dryer Design To form the basis for the analysis to characterize lint ignition, the operating characteristics of four different dryer designs were examined. The pattern of airflow; the locations of the heater, safety devices, and lint screen; and the tumbler and fan design were examined for each dryer design Airflow Pattern All four dryer designs had the same basic path for airflow. Air is pulled into the dryer through any gaps in the dryer housing, particularly through rear vents. Air is drawn over the heater, which warms the air, and then enters the tumbler. The air exits the tumbler and is directed through the lint screen. It then passes through a duct and into the fan. The fan forces the air through an exiting duct to the rear of the dryer. Figure 1 shows the airflow patterns for each of the dryer designs tested. Figure 1. Dryer Design Characteristics (Not to scale) 6

20 2.1.2 Heater Location and Configuration In all four clothes dryers, the heaters were located at the rear of the dryer. In three of the dryer designs, the heating element was contained in a rectangular metal enclosure, or heater box (Dryer Designs A, B and C). The location of the heater box varied among these three dryers. In two dryers, the heater boxes were oriented with the air intake directed towards the floor of the dryers (Dryer Designs A and B); and in the other dryer, the heater box was located near the top and the air intake directed towards the top of the dryer (Dryer Design C). The heating elements contained either one or two rows inside the heater box, as shown in Figure 2. The fourth dryer (Dryer Design D) had a circular heating element configuration as shown in Figure 3. The circular heating element was located behind the tumbler. Figure 2. Heating Element in a Heater Box Configuration (Does not represent any specific dryer design) Figure 3. Circular Heating Element in Dryer Design D (Not to Scale) Lint Screen Location In three of the four dryers, the lint screen was located at the front of the dryer and was accessible at the bottom of the door opening (Dryer Designs B, C and D). The lint screen of the fourth dryer was located at the rear of the dryer and was accessible from the top of the dryer (Dryer Design A) Tumbler Design The tumbler designs and sizes were similar for all four dryers with several noted differences. All the tumblers contained gaskets at the front and rear of the tumbler. Three of 7

21 the tumbler designs contained three baffles each that were spaced evenly. The other tumbler contained two baffles and a hump, evenly spaced Blower (Fan) Design All four dryers used a centrifugal-type fan. The centrifugal fan drew air through the center and forced it outward. All the fan blades were constructed of plastic or plastic reinforced with fibers Operating Features All dryer designs offered a choice of drying cycles: Timed Dry, Air Dry and Auto Dry. To eliminate any possible design variance in drying times during testing, only the Timed Dry cycle was used Safety Device Locations The locations of the safety devices on the dryers were similar. All four dryers had a minimum of two safety devices. One temperature switch (the primary or operating thermostat) was located after the lint screen and before the blower air intake. The second temperature switch (the high-limit thermostat) was located near the heater air intake. 2.2 Task 2: Document Dryer Operating Characteristics The main objective of this task was to record any similar or varying characteristics of a clothes dryer during normal and abnormal operations. The normal operation test was conducted with unrestricted airflow in the exhaust vent. Abnormal operation was defined as the dryer not cycling on the primary thermostat or cycling on the high-limit thermostat. The effects of a restricted exhaust vent were examined. The effects of a bypassed primary thermostat (to simulate thermostat failure) were also characterized Exhaust Vent Setup The same length and configuration of exhaust duct was used to vent each dryer design during testing. All venting material was 4-inch rigid metal duct. All joints were sealed with foil tape except the connection from the ducting to the dryer, which was secured using a 4-inch duct (hose) clamp. The exhaust vent was connected to a 4 angled wall cap. The wall cap contained a rodent screen with a grid of ¼-inch square openings. Figure 4 illustrates the setup used to vent the dryers. 8

22 Figure 4. Exhaust Vent Setup For tests in which airflow was restricted, an iris placed in series with the exhaust duct between the first and second 9-degree elbows was used to create a partially-blocked exhaust vent condition. Blockages of 25%, 5% and 75% were created using the iris. The percentage of blockage was calculated from that part of the cross section of the 4-inch duct that remained unblocked, as shown in Table 1. A blast plate covering 1% of the opening was used to create a fully-blocked exhaust vent (1% blockage). The blast plate was placed in the same location as the iris in the exhaust vent. Table 1. Unrestricted and Restricted Duct Configurations d (inches) Blockage (%) unrestricted 25 partially 5 partially 75 partially 1 blocked 9

23 Table 2 is a guide to determine the equivalent linear length of exhaust duct used in the testing. The table references were taken from Engineering and Design Clothes Dryer Exhaust Venting published by the Department of the Army, Army Corps of Engineers, 23 March 1998, No For every 9-degree elbow, the document specifies an equivalent length of 8 or 1 linear feet. Based on the calculated equivalent linear feet shown below, the exhaust vent configuration used in the test set up was within dryer manufacturer recommendations for installation for all dryers tested. Table 2. Equivalent Linear Duct Length Equivalent Linear Duct Length (feet) 8 equivalent for elbows 1 equivalent for elbows Elbow Straight Elbow Straight Wall Cap* 6 6 TOTAL * Conversion taken from dryer installation manuals Instrumentation Setup Each dryer was instrumented with six thermocouples and two hot wire anemometers. One additional thermocouple was used to record the ambient room temperature. All seven thermocouples were 24 gauge, K type. The 24 gauge thermocouple is relatively stiff and resulted in good stability in positioning the thermocouples. In addition, with a sampling rate of one sample per second, the response time of the thermocouples was adequate. The thermocouples were calibrated using a thermocouple oven before each dryer design was tested. The sampling rate during calibration was one sample every 1/1 second. Figure 5 shows the responses of the thermocouples versus the set oven temperature, recorded on November 14, 21, before Task 2 (Document Dryer Operating Characteristics) testing began. The figure shows a close match to the oven temperature except near the upper temperatures. 1

24 Thermocouple Calibration November 14, Temperature (C) Oven Temperature Data Points Oven temperature T1 T2 T3 T4 T5 T6 T7 Figure 5. Thermocouple Response Before Task 2 Testing At the end of this task, the thermocouples were re-checked, as shown in Figure 6. Figure 6 shows the high temperature closely follows the oven set temperature. The calibration/check tests taken in December 21 and early January 22 also have similar responses. Thermocouple Calibration January 22, 22 3 Temperature (C) Oven temperature T1 T2 T3 T4 T5 T6 T Oven Temperature Data Points Figure 6. Thermocouple Response After Task 2 Testing Figure 7 shows the average temperatures for all thermocouples compared to the oven temperatures for the days they were calibrated. The responses of the thermocouples show they are slightly below the oven set temperatures. The data presented in this document do not incorporate any correction factors in the thermocouple readings. 11

25 3 Thermocouple Calibration Comparison 25 2 Temperature (C) Oven Tempertuare Data Points Oven temperature November December 1 21 January 2 22 January Figure 7. Average Responses per Oven Set Temperatures and Day Table 3 lists the general locations of the thermocouples and anemometers in the dryers. Depending on the dryer design, the actual location may have varied slightly. Two hot wire anemometers were used to measure the airflow entering and exiting each dryer. One anemometer was placed at the intake into the heater, and the second anemometer was placed in the exhaust vent. Because of instrumentation limits, the sampling rate for the anemometers was one sample every 2 seconds. However, this was adequate since the analysis only needed to show trends in airflow. Table 3. General Locations of Thermocouples and Anemometers Thermocouple Number T1 T2 T3 T4 T5 T6 T7 Location Heater Intake Heater Exhaust Heater Housing Vent Exhaust Intake into Blower Intake into the Tumbler Ambient Room 12

26 Table 3-continued. General Locations of Thermocouples and Anemometers Anemometer Number V1 V2 Location Heater Intake Dryer Exhaust Global Test Procedure and Setup All dryers were tested at the high heat settings with no test load, with a dry load and with a wet load, unless otherwise specified. All dryers were tested with 25%, 5%, 75%, and 1% blockage in the exhaust vent with no test load and with a dry load, unless otherwise specified. Some of the clothes dryer instruction manuals specified a maximum load. The smallest of those was chosen as the test load. The standard test load was 1 bath towels of white 1% cotton terry, 45 long by 25.5 to 27 wide, each weighing approximately 57g (1.25 lbs.). For wet loads, the loads were washed and spun dry in a standard size washing machine set for Hot/Cold (hot wash/cold rinse) and Regular 1 [minutes]. No detergent was used in the wash. The clothes dryers were set for drying times of 6 minutes for wet loads and 15 minutes for dry loads using the Timed Dry feature. To accelerate the test process, the data for dry loads with no exhaust vent blockage was compared with data for wet loads with no exhaust vent blockage. If the thermocouple and airflow data were similar, additional tests with fully-blocked or partially-blocked exhaust vents were conducted using dry loads only (unless otherwise noted). The rodent screen was checked and cleaned after each set of dryer design tests. 13

27 2.2.4 Dryer Design A Dryer Design A was configured with the heater at the rear bottom of the dryer and the lint screen at the rear top of the dryer. The heated air entered one opening at the rear of the tumbler, and the moist air exited through a separate opening at the rear of the tumbler, as shown in Figure 8. Heated air into the tumbler Warm Moist Air exits the tumbler Exhaust Dryer Duct Intake Dryer Duct Inside the Tumbler Back of the Tumbler (rear cover removed) Figure 8. Dryer Design A Intake and Exhaust Dryer Design A Temperature and Airflow Characteristics, Blocked and Unblocked Exhaust Vent Conditions The dryer was tested at the high heat setting with no test load, with a dry load and with a wet load. Figure 9 (at the end of this section) shows a graph of the thermocouple and airflow measurement data for Dryer Design A with a wet load. (Similar graphs of data for the tests with no load and with a dry load are contained in Appendix A.) As can be seen in the graph, the dryer was started approximately 12 seconds after data collection began. The primary thermostat disconnected power to the heater approximately 3183 seconds into the test. It was observed that the exhaust vent airflow (V2) steadily decreased as the dryer operated. This was caused by the lint screen progressively becoming blocked with lint. Table 4 below lists the average, minimum, and maximum temperatures recorded during the main drying phase. The data listed in the table cover the period between 12 seconds after the dryer was started to 12 seconds before the primary thermostat removed power from the heating element (to eliminate any unstabilized readings from energizing and de-energizing the heating element). 14

28 Table 4. Dryer Design A Measurement Data ( C and sfpm) Temp Min Average Max Flow Min Average Max T V T V T T T T T The dryer was tested with a partially-blocked exhaust vent (25%, 5%, 75% blockage), and a 1% blocked exhaust vent. For these tests, dry loads were used, since the temperature data from the thermocouples were similar for both dry and wet loads for the unblocked condition. (Appendix A shows the comparison between the dry and wet towel loads with no blockage in the exhaust vent.) Figures 1 through 12 (at the end of this section) show a comparison of temperature measurement data for thermocouples T1, T3 and T2 for unblocked, partially-blocked, and fullyblocked vent conditions. (Appendix A contains graphs of the temperature data for the remaining thermocouples, T4, T5, T6 and T7.) In Figure 1, it can be seen that the dryer began to operate on the high limit thermostat only when the exhaust duct was fully (1%) blocked. For the 25% and 5% blocked exhaust vent conditions, the temperatures measured were similar to those measured for the unblocked exhaust vent condition. With a 75% blocked exhaust vent, the dryer still operated on the primary thermostat but at elevated temperatures; Figure 11 shows the heater box reached up to 15 C. In the unblocked condition or with exhaust vent blockages of 25% or 5%, the temperature of the heater box reached slightly over 1 C. At exhaust vent blockages of 1% and 75%, the peak heater exhaust temperatures were similar around 325 C to 375 C as shown in Figure 12. However, the period during which the heater exhaust operated near the peak temperature differed for the two conditions. At 1% blockage, the temperature was not maintained at the peak temperature very long because the high limit thermostat cycled the heating element more rapidly. At 75% blockage, the duration the heater exhaust stayed near the peak temperature was approximately 1 to 3 seconds until the primary thermostat switched the heating element off. 15

29 Dryer Design A Wet Load, High Heat Setting November 19, V2 T2 14 Temperature (C) V1 T1 T3 T7 T4 T6 T Air Flow (sfpm) Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Heater Housing T4 Exhaust Vent T5 Intake into Blower T6 Tumbler Intake T7 Room Ambient V1 Airflow into Heater V2 Airflow Exhaust Vent Figure 9. Dryer Design A Thermocouple and Airflow Measurements with a Wet Load 16

30 Dryer Design A T1 Heater Intake Temperature %, 25%, 5%, 75%, and 1% Blocked Dry Load, December Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 1. Dryer Design A T1 Heater Intake Comparison 17

31 Dryer Design A T3 Heater Housing Temperature %, 25%, 5%, 75%, and 1% Blocked Dry Load, December Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 11. Dryer Design A T3 Heater Housing Comparison 18

32 Dryer Design A T2 Heater Exhaust Temperature %, 25%, 5%, 75%, and 1% Blocked Dry Load, December Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 12. Dryer Design A T2 Heater Exhaust Comparison 19

33 Dryer Design A Airflow Airflow was measured using hot wire anemometers placed at the intake of the heater box and in the exhaust vent. Both anemometers were placed in the center of the cross-section of the air streams. The anemometer placed in the exhaust vent was positioned 36 after the first elbow to avoid circular turbulence effects from the dryer blower. As mentioned earlier, the iris and blast plate for the partially-blocked and 1% blocked conditions, respectively, were placed before the hot wire anemometer located in the exhaust vent. Figure 13 shows comparisons of intake air velocities when the exhaust vent was unblocked, partially blocked, and fully blocked, and the dryer contained a dry load. The intake velocities were similar for the unblocked and 25% and 5% blocked conditions. There was a slight drop in the air velocity for the 75% blocked condition, but it was not significant enough to cause the dryer to operate in the high-limit cycling mode (as can be observed from the thermocouple data in Figure 12). With a 1% blocked exhaust vent, the air intake velocity dropped to about 2 sfpm. Figure 14 is a graph comparing exhaust air velocities when the exhaust vent was unblocked, partially blocked, and fully blocked, and the dryer contained a dry load. The graph shows a slight overshoot in air velocity at initial startup of the dryer. The size of the overshoot decreased as the blockage increased. This was expected, since the anemometer was positioned after the blockage, which created a dampening effect. There was a slight increase in the velocity around 4 seconds, which was when the primary thermostat de-energized the heater. (This was also seen in Figure 9 Thermocouple and Airflow with a Wet Load, when the dryer began cycling.) This was caused by the hot wire anemometer not responding fast enough to the change in temperature. Figure 15 shows the average intake and exhaust air velocities for the different conditions of exhaust blockage. The data from 12 seconds to 1 seconds were averaged. This excluded the initial overshoot when the blower was first powered. The intake air velocity stayed fairly constant for the unblocked, 25% and 5% blocked conditions; however, it began to decrease more rapidly when the exhaust vent was 5% and 75% blocked. At 1% blockage, the exhaust air velocity was near zero, as expected, and the average intake velocity was 179 sfpm. Figure 16 is a graph similar to Figure 15 except that the dryer was operated with no test load and with no heat. The graph shows increased intake and exhaust air velocities when the tumbler was not filled as expected. Compared to Figure 15, there was a slightly more dramatic effect in the decrease of the air velocities as the exhaust vent was progressively blocked. 2

34 Dryer Design A Velocity Intake Comparison High Heat, Dry Load Velocity (sfpm) Time (seconds) No Block Intake 1% Blocked Intake 75% Blocked Intake 5% Blocked Intake 25% Blocked Intake Figure 13. Dryer Design A Intake Air Velocity 21

35 Dryer Design A Velocity Exhaust Comparison High Heat, Dry Load Velocity (sfpm) Time (seconds) No Block Exhaust 1% Blocked Exhaust 75% Blocked Exhaust 5% Blocked Exhaust 25% Blocked Exhaust Figure 14. Dryer Design A Exhaust Air Velocity 22

36 Dryer Design A Intake and Exhaust Velocity Average High Heat, Dry Load December Average Velocity (sfpm) (No Block) (Fully Blocked) Blockage (percentage) Intake Exhaust Figure 15. Dryer Design A Average Airflow Velocities (High Heat and Dry Load) 23

37 Dryer Design A Intake and Exhaust Velocity Average No Heat, No Load December 21 Average Velocity (sfpm) (No Block) (Fully Blocked) Blockage (percentage) Intake Exhaust Figure 16. Dryer Design A Average Airflow Velocities (No Heat and No Load) 24

38 Dryer Design A Primary Thermostat Bypassed The primary thermostat was bypassed to simulate a failure mode in which it failed closed. The thermostat was removed from the intake blower housing, and the hole it left was covered with foil tape. The thermostat was placed in a cool part of the dryer to prevent it from switching open. The dryer was operated with full airflow (no exhaust vent blockage) and a dry load. Figure 17 shows the temperature and airflow data for Dryer Design A with a bypassed thermostat. The dryer was switched to Air Dry at approximately 15 seconds into the test. The temperature at the heater intake appeared to level out at 95 C and would not have triggered the High-Limit Thermostat. 25

39 Dryer Design A Bypassed Primary Thermostst Dry Load December 1, V2 T2 Heater Element turned Off Switched to Air 15s Temperature (C) V1 T3 T4 T7 T1 T6 T Air Flow (sfpm) Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Heater Housing T4 Exhaust Vent T5 Intake into Blower T6 Tumbler Intake T7 Room Ambient V1 Airflow into Heater V2 Airflow Exhaust Vent Figure 17. Dryer Design A Primary Thermostat Bypassed 26

40 2.2.5 Dryer Design B Dryer Design B was configured with the heating element at the rear bottom of the dryer and the lint screen at the front of the dryer. The heated air entered the rear of the tumbler, and the moist air exited through an opening at the top of the lint screen, as shown in Figure 18. Heated air into the tumbler Warm Moist Air exits the tumbler Lint Screen Figure 18. Dryer Design B Air Flow Through the Tumbler Dryer Design B Temperature and Airflow Characteristics, Blocked and Unblocked The clothes dryer was tested at the high heat setting with no test load, with a dry load and with a wet load. Figure 19 (at the end of this section) shows a graph of the thermocouple and airflow measurement data for Dryer Design B with a wet load. (Graphs of data for similar tests with no load and with a dry load are contained in Appendix B.) As can be seen in the graph, the clothes dryer was started approximately 3 seconds after data collection began. The primary thermostat disconnected power to the heater approximately 3721 seconds into the test. It can be observed that the exhaust vent (V2) airflow significantly decreased over time as the dryer was operating. This was caused by the lint screen becoming progressively blocked with lint. This graph shows a larger drop in exhaust airflow than previously observed for Dryer Design A. In Dryer Design A, the exhaust air velocity dropped from approximately 13 to 95 sfpm, or a delta of 35 sfpm. In Dryer Design B, the exhaust air velocity dropped from approximately 14 sfpm to 8 sfpm, or a delta of 6 sfpm. Table 5 below lists the average, minimum, and maximum temperatures recorded during the main drying phase. The data listed in the table cover the period between 12 seconds after 27

41 the dryer was started to 12 seconds before the primary thermostat removed power from the heating element (to eliminate any unstabilized readings during energizing and de-energizing of the heating element). Table 5. Dryer Design B Statistics ( C and sfpm) Temp Min Average Max Flow Min Average Max T V T V T T T T T The dryer was tested with a partially-blocked exhaust vent (25%, 5%, 75% blockage) and 1% blocked exhaust vent. For these tests, wet towels (washed and spun dry) were used, since there was enough variation in the temperature data between dry and wet loads. (Appendix B shows the comparison between the dry and wet towel loads with no blockage in the exhaust vent.) Figures 2 to 22 (at the end of this section) show comparisons of temperature data for thermocouples T1, T3, and T5 for unblocked, partially-blocked, and fully-blocked vent conditions. (Appendix B contains graphs of the temperature data for the remaining thermocouples, T2, T4, T6, and T7.) In Figure 2, it can be seen that the dryer began to operate on the high limit thermostat only when the exhaust duct was either 75% or 1% blocked. For the 25% and 5% blocked exhaust vent conditions, the temperatures measured were similar to those measured for the unblocked exhaust vent condition. With a 75% and 1% blocked exhaust vent, the dryer operated on the high-limit thermostat. However, in the 75% blocked vent condition, the dryer operated at a higher temperature than that seen for the 1% blocked condition. Figure 21 shows the heater box reached up to 15 C for the 75% blocked condition whereas, for the 1% blocked condition, the high-limit thermostat periodically switched the temperature at approximately 125 C. The period at which the high-limit thermostat switched on and off also differed for the 75% and 1% blocked exhaust vent conditions. This was most likely caused by the difference in airflow for the two conditions. With the 75% blockage, the additional airflow resulted in an increase in the time required for the high-limit thermostat to reach its set point for opening. In addition, the periodic rate at which the thermostat switched on and off changed during the course of the test; the high limit thermostat began switching at a faster rate around 285 seconds. This was most likely caused by the lint screen becoming progressively blocked, which further reduced airflow through the dryer. Figure 2 also shows that the temperatures measured for the unblocked, 25% blocked and 5% blocked exhaust vent conditions are similar both in value and signature until

42 seconds. The graph for the 5% blocked condition shows the heating element began switching rapidly on the primary thermostat, which is explained in the next paragraph. Figure 21 shows that the temperature at which the operating thermostat switches on and off is nearly equal for the unblocked, 25% blocked and the 5% blocked conditions. The rapid switching of the primary thermostat was caused by the load no longer tumbling, although the drum was still rotating. (More on this phenomenon is discussed under Dryer Design D.) This effect was more evident for the 5% blocked exhaust condition, as shown in Figure

43 Dryer Design B Wet Load, No Blockage January 3, T Temperature (C) V1 T1 V2 T4 T5 T6 T7 T Air Flow (sfpm) Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Heater Housing T4 Exhaust Vent T5 Intake into Blower T6 Heater Exhaust T7 Room Ambient V1 Airflow into Heater V2 Airflow Exhaust Vent Figure 19. Dryer Design B Thermocouple and Airflow Measurements (Wet Load) 3

44 Dryer Design B T1 Heater Intake Temperature %, 25%, 5%, 75%, and 1% Blocked Wet Load January Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 2. Dryer Design B T1 Heater Intake Comparison 31

45 Dryer Design B T3 Heater Housing Temperature %, 25%, 5%, 75%, and 1% Blocked Wet Load January Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 21. Dryer Design B T3 Heater Housing Comparison 32

46 Dryer Design B T5 Intake into Blower Temperature %, 25%, 5%, 75%, and 1% Blocked Wet Load January Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 22. Dryer Design B T5 Intake into the Blower Comparison 33

47 Dryer Design B Airflow The airflow was measured using hot wire anemometers placed at the air intake into the heater box and in the exhaust vent. Both anemometers were placed in the center of the crosssection of the air streams. The anemometer placed in the exhaust vent was positioned 36 after the first elbow to avoid circular turbulence effects from the dryer blower. As mentioned earlier, the iris and blast plate for the partially and 1% blocked conditions, respectively, were placed before the hot wire anemometer. Figure 23 shows the intake air velocity comparisons for the unblocked, partially blocked, and fully blocked exhaust vent conditions. The intake air velocities were similar for the unblocked, 25% and 5% blocked conditions. There was a slight drop in the velocity for the 75% blocked condition that was significant enough to cause the dryer to operate in the high-limit cycling mode, as seen in the thermocouple data (shown previously in Figure 21). When the exhaust vent was 1% blocked, the intake air velocity fluctuated between near zero to about 2 sfpm. The fluctuation to 2 sfpm coincided with the heating element being de-energized by the high-limit thermostat. This was caused by the anemometer not responding fast enough to the change in temperature at the heater intake. Figure 24 shows the exhaust air velocity comparisons for the unblocked, partially blocked, and fully blocked exhaust vent conditions. The graph shows a slight overshoot at initial startup of the dryer, but it was not as prominent as that seen in Dryer Design A. The size of the overshoot decreased as the blockage increased, as expected, since the second anemometer, in the exhaust vent, was located after the iris. Figure 25 shows the average intake and exhaust air velocities for the different blockage conditions. The data from 12 seconds to 1 seconds were averaged. This avoided the initial overshoot when the blower was first powered. The intake velocity stayed steady until the exhaust vent was 25% blocked, and it began to decrease more rapidly between 25% and 5% blockage. When the exhaust vent was 1% blocked, the exhaust velocity was near zero, as expected, and the average intake air velocity was 9 sfpm. 34

48 Dryer Design B Velocity Intake Comparison High Heat, Wet Load Velocity (sfpm) Time (seconds) No Block Intake 1% Blocked Intake 75% Blocked Intake 5% Blocked Intake 25% Blocked Intake Figure 23. Dryer Design B Intake Air Velocity 35

49 Dryer Design B Velocity Exhaust Comparison High Heat, Wet Load Velocity (sfpm) Time (seconds) No Block Exhaust 1% Blocked Exhaust 75% Blocked Exhaust 5% Blocked Exhaust 25% Blocked Exhaust Figure 24. Dryer Design B Exhaust Vent Air Velocity 36

50 Dryer Design B Intake and Exhaust Velocity Average High Heat, Wet Load January 22 Average Velocity (sfpm) (No Block) (Fully Blocked) Blockage (percentage) Intake Exhaust Figure 25. Dryer Design B Average Airflow Velocities (High Heat and Wet Load) 37

51 Dryer Design B Primary Thermostat Bypassed The primary thermostat was bypassed to simulate a failure mode in which it had failed closed. The thermostat was removed from the intake blower housing, and the hole it left was covered with foil tape. The thermostat was placed in a cool part of the dryer to prevent it from switching open. The dryer was operated with full airflow and with a wet load. Figure 26 shows the thermocouple and airflow measurement data for Dryer Design B when the thermostat was bypassed. The dryer s Timed Dry feature shut off at approximately 4 seconds. The dryer was restarted for an additional 3 minutes with the Timed Dry feature. The temperature at the heater box intake appeared to level out at approximately 65 C and would not have triggered the High-Limit Thermostat. 38

52 Dryer Design B Bypassed Primary Thermostst Wet Load February 11, 22 4 Dryer Timed Setting Switched Off V Temperature (C) T2 T6 Reset Timed Setting for 3 minutes Switched Off Air Flow (sfpm) 1 V1 4 5 T3 T4 T5 T T1 2 Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Heater Housing T4 Exhaust Vent T5 Intake into Blower T6 Heater Exhaust T7 Room Ambient V1 Airflow into Heater V2 Airflow Exhaust Vent Figure 26. Dryer Design B Primary Thermostat Bypassed 39

53 2.2.6 Dryer Design C Dryer Design C was configured with the heating element at the rear-top of the dryer and the lint screen at the front of the dryer. The heated air entered the rear of the tumbler, and the moist air exited through openings inside the front cover that led to the lint screen, as shown in Figure 27. Heated air into the tumbler Warm Moist Air exits the tumbler Lint Screen Inside View of Front Cover Figure 27. Dryer Design C Airflow through the Tumbler Dryer Design C Temperature and Airflow Characteristics, Blocked and Unblocked The dryer was tested at the high heat setting with no load, with a dry load and with a wet load. Figure 28 shows the thermocouple and airflow measurement data for Dryer Design C with a wet load. (Graphs of data obtained for the tests with no load and with a dry load are contained in Appendix C.) The dryer was started approximately 6 seconds after data collection began. The primary thermostat disconnected power to the heating element at approximately 3664 seconds into the test. This graph shows a significant decrease in the exhaust air velocity as the dryer continued to operate. As seen in Dryer Designs A and B, this was caused by the lint screen becoming progressively blocked with lint. In Dryer Design C, the exhaust air velocity dropped from approximately 14 sfpm to 7 sfpm, a delta of 7 sfpm. Table 6 below lists the average, minimum, and maximum temperatures recorded during the main drying phase. The data listed in the table cover the period from 12 seconds after the dryer was started to 12 seconds before the primary thermostat removed power from the 4

54 heating element (to eliminate any instability in the readings from energizing and de-energizing the heating element). Table 6. Dryer Design C Statistics ( C and sfpm) Temp Min Average Max Flow Min Average Max T V T V T T T T T The dryer was tested with a partially-blocked exhaust vent (25%, 5% and 75% blockage) and a 1% blocked exhaust vent. For these tests, wet towels (washed and spun dry) were used. (A comparison of temperature and airflow data obtained for the dry and wet towel loads with no blockage in the exhaust vent are contained in Appendix C.) Figures 29 through 31 (at the end of this section) show a comparison of the temperature measurement data for thermocouples T1, T2, and T5 for unblocked, partially-blocked, and fullyblocked vent conditions. (Appendix C contains graphs of the temperature measurement data for unblocked, partially-blocked, and fully-blocked vent conditions for the remaining thermocouples, T3, T4, T6, and T7.) Figure 29 shows that the dryer operated on the high-limit thermostat only when the exhaust duct was 75% or 1% blocked. For the 25% and 5% blocked exhaust vent conditions, the temperatures measured were similar to those measured when the exhaust vent was unblocked. Figure 29 also shows that, with 75% and 1% exhaust vent blockages, there appeared to be a difference in the temperature at which the high-limit thermostat operated. At 75% blockage, upper high-limit switching occurred around 195 C; however, for a 1% blocked exhaust vent, upper high limit switching appeared to have occurred at 225 C. This discrepancy in the data was caused by the location of the T1 thermocouple and the orientation of the heater box. As mentioned previously, Dryer Design C drew intake air into the heater box from the top and directed it downward toward and eventually into the tumbler. When the exhaust vent was 75% blocked, there was slight airflow over the heating element to direct some heat away from the T1 thermocouple and the high-limit thermostat. With 1% blockage, convective heat caused the T1 thermocouple to read a higher value a higher temperature than was actually present at the high-limit thermostat when it opened. Figure 3 shows that when the exhaust vent was either partially or fully blocked, the heater housing temperature was only slightly above the maximum temperature for the unblocked vent condition. The operation of the high-limit thermostat prevented the heater box temperature from increasing much higher than 8 C. 41

55 Figure 31 shows the air temperature at the intake to the blower reached around 5 C in the unblocked and partially-blocked conditions, until around 21 seconds. For the unblocked, 25% and 5% blocked conditions, the temperature continued to increase to over 6 C. With the exhaust vent 75% blocked, switching of the high-limit thermostat prevented the temperature of the air entering the blower from increasing to the set point at which the primary thermostat would open. 42

56 Dryer Design C High Heat, Wet Load December 11, V2 T Temperature (C) 15 1 T Air Flow (sfpm) 5 T1 T3 T4 T T7 V1 2 Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Heater Housing T4 Exhaust Vent T5 Intake into Blower T6 Heater Exhaust T7 Room Ambient V1 Airflow into Heater V2 Airflow Exhaust Vent Figure 28. Dryer Design C Thermocouple and Airflow Measurements (Wet Load) 43

57 Dryer Design C T1 Heater Intake Temperature %, 25%, 5%, 75%, and 1% Blocked High Heat, Wet Load December 21 Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 29. Dryer Design C T1 Heater Intake Comparison 44

58 Dryer Design C T3 Heater Housing Temperature %, 25%, 5%, 75%, and 1% Blocked High Heat, Wet Load December Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 3. Dryer Design C T3 Heater Housing Comparison 45

59 Dryer Design C T5 Intake into Blower Temperature %, 25%, 5%, 75%, and 1% Blocked High Heat, Wet Load December 21 Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 31. Dryer Design C T5 Intake into the Blower Comparison 46

60 Dryer Design C Airflow The airflow was measured using hot wire anemometers placed at the intake airflow into the heater box and in the exhaust vent. Both anemometers were placed in the center of the cross-section of the air streams. The anemometer placed in the exhaust vent was positioned 36 after the first elbow to avoid circular turbulence effects from the dryer blower. As mentioned earlier, the iris and blast plate for the partially- and fully-blocked conditions, respectively, were placed before the hot wire anemometer located in the exhaust vent. Figure 32 shows the intake air velocity comparisons for unblocked, partially-blocked, and fully-blocked exhaust vent conditions. The intake air velocity appeared to have very low readings; this was caused by the presence of side louvers in the heater box. The anemometer did not measure the air intake from the side louvers. As seen previously when the exhaust vent was 1% blocked, the intake air velocity fluctuated between near zero to about 2 sfpm. The fluctuation to 2 sfpm coincided with the heating element being de-energized by the high-limit thermostat. This was caused by the anemometer not responding fast enough to the temperature change when the heater was de-energized. Figure 33 shows the exhaust air velocity comparisons for unblocked, partially-blocked, and fully-blocked exhaust vent conditions. The graph shows a slight overshoot at initial startup of the dryer, but it is not as prominent as that seen for Dryer Design A. The size of the overshoot decreased as the blockage increased, as expected, since the second anemometer was located after the exhaust vent blockage. The exhaust air velocity decreased when the exhaust vent was 25% and 5% blocked, and it decreased further when the exhaust vent was 75% blocked. Figure 34 shows the average intake and exhaust air velocities for the different blockage conditions. The data from 12 seconds to 1 seconds were averaged. This avoided the initial overshoot when the blower was first powered. 47

61 Dryer Design C Velocity Intake Comparison High Heat, Wet Load Velocity (sfpm) Time (seconds) No Block Intake 1% Blocked Intake 75% Blocked Intake 5% Blocked Intake 25% Blocked Intake Figure 32. Dryer Design C Intake Air Velocity 48

62 Dryer Design C Velocity Exhaust Comparison High Heat, Wet Load Velocity (sfpm) Time (seconds) No Block Exhaust 1% Blocked Exhaust 75% Blocked Exhaust 5% Blocked Exhaust 25% Blocked Exhaust Figure 33. Dryer Design C Exhaust Vent Air Velocity 49

63 Dryer Design C Intake and Exhaust Velocity Average High Heat, Wet Load December Average Velocity (sfpm) (No Block) (Fully Blocked) Blockage (percentage) Intake Exhaust Figure 34. Dryer Design C Average Airflow Velocities (High Heat and Wet Load) 5

64 Dryer Design C Primary Thermostat Bypassed The primary thermostat was bypassed to simulate a failure mode in which it failed closed. The thermostat was removed from the intake blower housing, and the hole it left was covered with foil tape. The thermostat was placed in a cool part of the dryer to prevent it from switching open. The dryer was operated with full airflow (no exhaust blockage) and with a dry load. Figure 35 shows the thermocouple and airflow measurement data for Dryer Design C with a bypassed primary thermostat. The graph shows that the dryer began operating on the high-limit thermostat. Since evaluations of Dryer Designs A and B did not result in high-limit cycling when the primary thermostat was bypassed, inspection for a blocked exhaust vent was conducted. It was determined that lint on the rodent screen at the vent hood reduced airflow sufficiently to cause the dryer to operate in the high-limit cycling mode. The lint (approximately.69 grams and distributed evenly) was removed at approximately 2 seconds into the test. For the remainder of the test, the dryer no longer cycled on the high-limit thermostat. 51

65 Switching on the High-Limit Thermostat Dryer Design C Primary Thermostat Bypassed Dry Load January 2, Cleaned Blocked Exhaust 14 2 T Temperature (C) 15 1 V2 V1 T2 T1 T4 T Air Flow (sfpm) 5 T T7 2 Time (seconds) T1 Heater Intake T2 heater Exhaust (3.5) T3 Heater Housing T4 Exhaust Vent T5 Intake into Blower T6 Heater Exhaust (5.5) T7 Room Ambient V1 Airflow into Heater V2 Airflow Exhaust Vent Figure 35. Dryer Design C Primary Thermostat Bypassed 52

66 2.2.7 Dryer Design D Dryer Design D was configured with the heating element behind the tumbler and the lint screen at the front of the dryer. The heated air entered the rear of the tumbler, and the moist air exited through openings at the rear of the front cover leading to the lint screen, as shown in Figure 36. Heated air into the tumbler Warm Moist Air exits the tumbler Lint Screen Inside View of Front Cover Figure 36. Dryer Design D Airflow Through the Tumbler Dryer Design D Temperature and Airflow Characteristics, Blocked and Unblocked Limited testing of this clothes dryer at the high heat settings with no load, with a dry load and with a wet load was conducted. At the end of the wet load testing (6 minutes on Timed Dry), the load was still very damp and had two different cycling periods as shown later. To determine why the heater had two different cycling periods and the load was still damp at the end of the test, a relay and a 9-volt battery were placed in parallel with the heating element and the high-limit thermostat, as shown in Figure 37. The T2 thermocouple was replaced with the signal from the 9-volt battery. The relay monitored the voltage between the primary and high-limit thermostats. If both thermostats were in their normally-closed positions, the relay would apply 9-volts (actually 7 volts) to the T2 input of the data acquisition system. If the primary thermostat opened, the relay would close and the signal to the data acquisition system would be near zero. If the high-limit thermostat opened, the relay would still be powered and a 9-volt signal would be recorded on the data acquisition system and the thermocouples would indicate that the heating element was cooling down. 53

67 24 VAC 12 VAC Heater Element 12 VAC Primary Thermostat High-Limit Thermostat To Data Acquisition (T2) 9-volt battery Relay Figure 37. Setup with a Relay and Battery for T2 Analog Figure 38 shows the thermocouple and airflow measurement data for Dryer Design D with a wet load before setup, as shown in Figure 37. The primary thermostat first disconnected power to the heater at approximately 779 seconds into the test. The heater cycled approximately every 374 seconds, until 3133 seconds into the test. After 3133 seconds, it began to cycle approximately every 19 seconds. At the end of the test 6 minutes on Timed Dry the load was still very damp (Lists of the tests and additional figures for Dryer Design D are located in Appendix D.) The standard load of 1 towels was tested again, but with the T2 analog signal/relay. Figure 39 shows the thermocouple, airflow, and T2 analog data for a wet load. The graph shows that the dryer behaved the same as it had in the previous test; i.e., it exhibited the phenomenon of two periodic cycles for the thermostats. In this case, however, it could be determined from the T2 signal that the first periodic cycling was caused by the high-limit thermostat, and the second periodic cycling was caused by the primary thermostat. As before, the towels were still very damp at the end of the 6-minute (Timed Dry) drying cycle. The most likely cause for the dryer to begin operating on the high-limit thermostat in the unblocked exhaust vent condition was inadequate airflow through the tumbler. The dryer appeared to be responding to an overload condition in which the load restricted the airflow through the tumbler. To determine why the primary thermostat began cycling later in the test, the dryer door was removed and replaced with a sheet of rigid clear plastic. It was observed that, after a period, the load stopped tumbling and began to ride along the sides of the rotating drum. This explained why the dryer began cycling on the primary thermostat. When the load was riding on the rotating drum, this allowed the heated air to pass freely through the tumbler and into the lint 54

68 screen/blower. The primary thermostat was able to detect the increase in temperature, causing it to open or cycle the heater off. When the tumbler temperature cooled, the primary thermostat closed or cycled the heater on. This also explained why the load was still damp at the end of the 6-minute drying cycle. This same phenomenon occurred when the dryer was tested with wet or damp loads of 1, 7, or 5 towels, as shown in Appendix D. (A video clip demonstrating this phenomenon is included in Appendix J.) This phenomenon also occurred with Dryer Design B, but it was not as repeatable as it was with Dryer Design D. The thermocouple and airflow measurement data for Dryer Design D are not listed here. The data would not be meaningful because of the operating characteristics of the dryer; i.e., immediate high-limit cycling due to overloading and the phenomenon of the load riding along the sides of the rotating drum. To obtain steady and consistent data, the dryer was tested with no load with a partiallyblocked exhaust vent (25%, 5%, 75% blockage), and a 1% blocked exhaust vent. Figures 4 and 41 show comparisons of temperature measurement data for unblocked, partially-blocked, and fully- blocked exhaust vent conditions for T1 and T3 thermocouples. (Appendix D contains graphs of the temperature data for the remaining thermocouples, T2, T4, T5, T6, and T7.) Figure 4 shows that the dryer began to operate on the high-limit thermostat when the exhaust duct was 75% or 1% blocked. For the 25% and 5% blocked exhaust vent conditions, the temperatures measured were similar to those measured when the exhaust vent was unblocked. For the 75% and 1% exhaust vent blockage conditions, the graph shows two different periodic rates at which the high-limit thermostat cycled. With the exhaust vent 75% blocked, the lower set point temperature for the high-limit thermostat was 5 C; when the exhaust vent was 1% blocked, it appeared to be switching at 1 C. It could not be determined why the high-limit thermostat was cycling at two different lower set points for the 75% and 1% blocked conditions. Figure 41 shows comparisons of temperature data at the heater box T3 thermocouple, for unblocked, partially-blocked, and fully blocked exhaust vent conditions. The housing temperature increased to approximately 5 C for the unblocked, 25% and 5% blocked vent conditions. At blockages of 75% and 1%, the housing temperature increased to approximately 15 C. 55

69 Dryer Design D High Heat, Wet Load January 3, V2 T Temperature (C) T6 T3 T Air Flow (sfpm) 5 T7 T4 T V1 4 2 Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Heater Housing T4 Exhaust Vent T5 Intake into Blower T6 Tumbler Intake T7 Room Ambient V1 Airflow into Heater V2 Airflow Exhaust Vent Figure 38. Dryer Design D Thermocouple and Airflow Measurements (Wet Load) 56

70 Dryer Design D High Heat, Wet Load, with Analog (T2) January 7, V2 T Temperature (C) T Air Flow (sfpm) 5 T1 T7 T2 analog T4 T V1 4 2 Time (seconds) T1 Heater Intake Analog T3 Heater Housing T4 Exhaust Vent T5 Intake into Blower T6 Tumbler Intake T7 Room Ambient V1 Airflow into Heater V2 Airflow Exhaust Vent Figure 39. Thermocouple, Airflow and Relay (T2 Analog) Measurements 57

71 Dryer Design D T1 Heater Intake Temperature %, 25%, 5%, 75%, and 1% Blocked High Heat, No Load January Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 4. Dryer Design D T1 Heater Intake Comparison 58

72 Dryer Design D T3 Heater Housing Temperature %, 25%, 5%, 75%, and 1% Blocked High Heat, No Load January Temperature (C) Time (seconds) No Block 75% Blocked 5% Blocked 25% Blocked 1% Blocked Figure 41. Dryer Design D T3 Heater Housing Temperature Comparison 59

73 Dryer Design D Airflow The airflow was measured using hot wire anemometers placed at the intake airflow into the heater and in the exhaust vent. The anemometer at the exhaust vent was placed in the center of the cross-section of the air stream. The other anemometer was placed at top-center of the circular heater intake. The anemometer placed in the exhaust vent was positioned 36 after the first elbow to avoid circular turbulence effects from the dryer blower. As mentioned earlier, the iris and blast plate for the partially-blocked and fully-blocked conditions, respectively, were placed before the hot wire anemometer located in the exhaust vent. Figure 42 shows the intake air velocity comparisons for the unblocked, partially-blocked, and fully-blocked exhaust vent conditions with no load. Due to the configuration of the heating element, the intake air velocity measurement data were not as accurate as the exhaust air velocity data. The air intake was configured in a 36 ring behind the tumbler, with a long narrow slot. The shape and surface area of the heater intake caused the airflow velocity to be below the resolution of the anemometer instrumentation. Figure 43 shows the exhaust air velocity comparisons for the unblocked, partiallyblocked, and fully-blocked exhaust vent conditions with no load. The graph shows a slight overshoot at initial startup of the dryer, which was caused by the blower being energized, as seen with the other dryer designs. There were periodic increases in exhaust airflow, which corresponded to the heating element cycling off when the primary thermostat opened. The increase in exhaust airflow was caused by the anemometer responding slowly to the sudden temperature changes. Figure 44 shows the average intake and exhaust air velocities for the different blockage conditions. The data from 12 seconds to 1 seconds were averaged. This would avoid the initial overshoot when the blower was first powered. The 25% and 5% blocked exhaust conditions displayed a more pronounced, higher exhaust airflow velocity than did the unblocked condition; this was not seen during examination of the other dryer designs. 6

74 Dryer Design D Velocity Intake Comparison High Heat, No Load Velocity (sfpm) Time (seconds) No Block Intake 1% Blocked Intake 75% Blocked Intake 5% Blocked Intake 25% Blocked Intake Figure 42. Dryer Design D Intake Air Velocity 61

75 Dryer Design D Velocity Exhaust Comparison High Heat, No Load Velocity (sfpm) Time (seconds) No Block Exhaust 1% Blocked Exhaust 75% Blocked Exhaust 5% Blocked Exhaust 25% Blocked Exhaust Figure 43. Dryer Design D Exhaust Vent Air Velocity 62

76 Dryer Design D Intake and Exhaust Velocity Average High Heat, No Load January Average Velocity (sfpm) (No Block) (Fully Blocked) Blockage (percentage) Intake Exhaust Figure 44. Dryer Design D Average Airflow Velocities (High Heat and No Load) 63

77 Dryer Design D Primary Thermostat Bypassed The primary thermostat was bypassed to simulate a failed closed primary thermostat. The thermostat was removed from the intake blower housing, and the hole was covered with foil tape. The thermostat was placed in a cool part of the dryer to prevent it from switching open. The dryer was operated with full airflow (no exhaust blockage) and a wet (washed and spun dry) load. To avoid the phenomenon where the load stopped tumbling after a period, the dryer load was reduced to six cotton towels. Five of the six towels were of the same size as used before. The sixth towel was a different size, approximately 25% larger, to help prevent the load from riding the tumbler. The T2 analog signal was used to determine if either the primary or the highlimit thermostat opened. Figure 45 shows the thermocouple and airflow measurement data for Dryer Design D with a bypassed primary thermostat. The load began to ride the sides of the tumbler at approximately 19 seconds after the dryer was started. Approximately 3 seconds after the dryer was started, the high-limit thermostat opened, as shown by the graph of the T2 analog signal. For unknown reasons, the T1 thermocouple became very erratic just before the highlimit switch opened. The test was stopped at approximately 325 seconds after the dryer was started. 64

78 Dryer Design D Wet Load (6 towels) January 16, 22 Temperature (C) V1 T V2 The Load stops Tumbling T3 T6 High-Limit Switch Opens Power Turned Off Switch Opens Air Flow (sfpm) T5 T4 5 T1 T2 T Time (seconds) 4 2 T1 Heater Intake T2 Analog T3 Heater Housing T4 Exhaust Vent T5 Intake into Blower T6 Heater Exhaust T7 Room Ambient V1 Airflow into Heater V2 Airflow Exhaust Vent Figure 45. Dryer Design D Primary Thermostat Bypassed 65

79 2.3 Task 3: Monitor Lint Distribution The objective of this task was to monitor lint distribution and accumulation in areas within a clothes dryer during operation with normal airflow. Lint accumulation was characterized in one dryer, and an analysis of potential causes for lint accumulation in other dryer designs was conducted Test Setup Dryer Design A was selected for testing. It provided the most convenient access to the interior of the dryer without removing or disturbing any interior dryer components. Two view ports of clear plastic were installed on opposite sides of the dryer chassis (one is shown in Figure 46(a)). One view port measured 3.5 inches by 19.5 inches; the other measured 3.5 inches by 23 inches. A silicone-type caulk was used to seal the edges of the view ports. During testing, the back panel of the dryer was installed; however, it was temporarily removed, as shown in Figure 46(b), to inspect the lint accumulation before, during, and after testing. (a) (b) Figure 46. Dryer Design A Used in Task 3 Testing The dryer was vented to the outside of the building. All venting material was 4-inch rigid metal duct. All joints were sealed with foil tape except for the connection to the dryer; the exhaust vent was secured to the dryer using a 4-inch duct (hose) clamp. 66

80 2.3.2 Test Description For this task, loads of towels were alternately washed in detergent-free hot water and spun dry, then dried for approximately 6 minutes in the dryer using the Timed Dry feature (one washing followed by drying equals one cycle). A total of 1 cycles was completed. After 48 and 1 cycles, the interior of the dryer was inspected and the presence of any lint accumulation was documented. Four loads of towels were used in Task 3 testing; a load consisted of 1 white cotton towels. During the first 48 cycles, two of the four loads were used for the test. For cycles 49 through 1, the remaining two loads were used. Before each cycle, a load was conditioned overnight (for approximately twelve hours) in a conditioning chamber with a controlled relative humidity (RH) of 5 ± 5% and a temperature of 23 to 25 C. Two test methods to record the amount of lint accumulation in the dryer were attempted. Method 1 was to record the weight loss of a load resulting from the washing and drying cycles. Method 2 was to visually inspect the interior of the dryer for lint accumulation Method 1 Weight Loss of Load Method 1 was used to estimate the amount of towel material lost (as lint) into the interior of the test dryer by recording the weight loss of a load resulting from the washing and drying cycles. The weight of a conditioned load was recorded before the start of a cycle. The amount of towel material lost during washing, the amount that accumulated on the lint screen during drying, and the amount lost through the dryer vent were recorded. The remainder of any lost material was assumed to have escaped into the interior of the dryer chassis. This method proved to be unsuitable for two reasons. The first was the fluctuation in the humidity level in the conditioning chamber, which varied by ± 5% RH. The second was the rate at which the towels lost moisture outside the conditioning chamber, depending on the humidity. To determine the weight variation in the conditioned load due to the fluctuation in relatively humidity of the conditioning chamber, a load of towels was stored overnight in a room with approximately 25% RH. The load was then transferred to the conditioning chamber. After conditioning for a minimum of 14 hours, the towel load was weighed. The consequent weight gain was approximately 1g, or 4g per 1% RH. With the conditioning chamber having a variation of ± 5% RH, the weight of a 5g load could vary by ± 2g. The difference in humidity between the conditioning chamber and the test room caused the weight of the towels to vary more significantly than the amount of lint lost. The towels were weighed in the test room after they were removed from the conditioning chamber. During the time that the towels were weighed, moisture in the towels could evaporate. The relative humidity levels in the test room and the conditioning chamber were 2-3% RH and 5 ± 5% RH, respectively. The average weights of lint samples collected during one cycle were.6g (after wash),.4g (in the lint screen), and.2g (in the dryer vent), or a combined weight of 1.2g. The 67

81 uncertainty in the weight of a load ( ± 2g) due to changes in humidity was much greater than measurable lint loss due to the washing and drying cycle. It can also be assumed that the uncertainty in the weight of a load would be even greater than the measurable lint lost in the dryer chassis interior per cycle. The unsuitability of this test method was further verified by calculating the standard deviation of the load weight before wash. The standard deviation was ± 1g, and the standard deviation of the mean was ± 2g. In order to detect a loss of lint during the drying process, it would be necessary to resolve a weight differential of less than 1.g and a relative humidity of 1%. This was not possible with the available test conditions (load weight varying by ± 2 g and fluctuation of ± 5% RH in the conditioning chamber) Method 2 Visual Examination Method 2 was to visually observe any accumulation of lint inside the dryer chassis before, during, and after testing. After 48 cycles (using the first two loads of towels), a preliminary count of accumulated lint and dust particles was performed. Some lint accumulation, as well as some towel degradation, was observed. The lint accumulation test continued for another 52 cycles using the second two loads of new towels. A total of 1 cycles was completed. After 1 cycles, visible evidence of lint accumulation was detected (as shown in Figure 47). A lint ball approximately one inch in diameter was observed inside the chassis. The lint was white and, therefore, it was inferred that it was from the towel material. There appeared to be no significant admixture of room dust, which was not likely to have been white in color. Figure 47. Lint Accumulated at Bottom of Dryer After 1 cycles Figure 48 shows lint that accumulated on the interior dryer vent after 48 cycles. Figure 49 shows the same area after 1 cycles. An increase in the number of lint particles was noticeable. 68

82 Figure 48. Lint Accumulated on Interior of Dryer After 48 Cycles Figure 49. Lint Accumulated on Interior of Dryer After 1 Cycles (same location as shown in Figure 48) The number of lint particles in an approximately half-inch square shown in Figures 48 and 49 was estimated. Approximately 83 particles were counted after 48 cycles; there were approximately 25 particles after 1 cycles. Some areas of the dryer did not show an observable gain in lint coverage, while other areas had significant gains Examination of Dryer Design A Figure 5 shows an area of the blower fan housing after 1 cycles. A significant accumulation of lint was observed at a sealed joint for the blower housing (shown by white arrow), and a lint ball was observed on the dryer floor (shown by black arrow). This accumulation was not apparent at the start of Task 3 testing or after 48 cycles. 69

83 Figure 5. Lint Build-up at Blower Housing Seal and Dryer Floor 7

84 2.4 Task 4: Determine Characteristics Required for Lint Ignition The main objective of this task was to record the ignition characteristics of lint on, near, and ingested into the heater. The test variables used in the Task 4 testing were developed from data collected in Task 2. Document Dryer Operating Characteristics. The testing was divided into two phases. Phase I consisted of examining the ignition characteristics of lint on and near the heater. Phase II consisted of examining the ignition characteristics of lint that was ingested into the heater Test Setup A test apparatus was constructed to house the heater box/element (from Dryer Design A) and venting material for the heater exhaust, as shown in Figures 51 and 52. The test setup allowed precise control of the test conditions. Two fans and irises were installed to independently control the airflow external and through the heater box. The heater was wired to a variable transformer to control the heat output, if needed. All venting material was 4-inch rigid metal duct. All joints were sealed with foil tape. Appendix F illustrates the dimensions of the test setup used in Task 4. Although a heater from one of the dryer designs was used, the test setup removed many of the design variables found in any particular dryer design, and staff believes the test results may be applied to the other dryer designs Instrumentation Setup The test setup was instrumented with six thermocouples and two hot wire anemometers. One additional thermocouple was used to record the ambient room temperature. All seven thermocouples were 24 gauge, K type. The same thermocouples used in the Task 2 testing were used in this phase of testing. Table 6 lists the locations of the thermocouples. The T3 thermocouple was cemented to the heater housing using a ceramic based cement. Table 6. Thermocouple Location and Setup Thermocouple Location T1 Heater Intake T2 Heater Exhaust T3 Heater Housing or Lint Sample* T4 Intake into Blower/Heater Exhaust T5 Intake into Blower/Outside Heater T6 Chamber Temperature T7 Ambient Room * Location may differ, as specified in the report 71

85 Two anemometers were used in this phase of testing. One anemometer was placed at the heater box intake. The second anemometer was placed near the top of the heater box to measure the airflow over the heater box housing. 72

86 Heater Heater Exhaust Inlet View Ports Figure 51. Test Setup for Lint Ignition 73

87 Heater Box Airflow Direction Anemometer Figure 52. Test Setup for Lint Ignition 74

88 2.4.3 Phase I Ignition Characteristics of Lint On and Near the Heater Box In this phase of testing, lint samples were placed on top of the heater housing, at one side of the heater housing, and at the heater intake. All lint samples were conditioned for a minimum of 24 hours at 23 C to 25 C and 5% to 55% relative humidity. The lint samples were tested at five different locations on the heater housing as shown in Figure 53. Lint samples were tested in two locations on top of the heater housing, locations A and B. Lint samples were tested at three locations on one side of the housing, locations C, D and E. Airflow Direction Figure 53. Location of Lint Samples on the Housing Top of Heater Housing The first series of tests was conducted with lint samples placed on top of the heater housing at location A, as shown in Figure 54. The high-limit thermostat was bypassed for this series of tests. The lint samples measured approximately 2 x 2 x ¼ inches and weighed approximately.3 grams. The samples were 1% cotton lint collected from the lint screen during Task 3 testing. 75

89 T3 Thermocouple Lint Sample Figure 54. Lint Sample on Heater Housing at Location A Table 7 provides a list of the lint ignition tests that were conducted at location A on top of the heater housing. The table includes the weights of the lint samples tested, test conditions, and the observed results. Location Number Weight (grams) Table 7. Testing on Top of the Heater Housing at Location A Power Intake Airflow (sfpm) Outside Airflow (sfpm) High-Limit Thermostat Result A s step bypassed smolder A s step bypassed smolder A s step 4 bypassed smolder A s instant 4 bypassed smolder A s1-5a.32 instant 4 bypassed smolder A s2-8.3 instant 4 bypassed smolder A s instant 4 bypassed smolder The samples either smoldered or charred during the tests, as shown in Figure 55. In the first three tests (Sample Numbers s1-1, s1-2, and s1-3), the power to the heater was incrementally increased. In the next three tests (s1-4, s1-5a, s2-8), power was applied instantaneously to the heating element. In all these tests, the samples smoldered but did not ignite. In test s5-1, the airflow through the heater was initially 65 sfpm until 4 seconds when it was decreased to zero sfpm. The sample charred and smoldered. 76

90 Sample 1-5a Sample 1-5a Before Smoldering Figure 55. Sample Smoldering at Location A Table 8 provides a list of the lint ignition tests that were conducted at location B on top of the heater housing. The lint samples measured approximately 2 x 2 x ¼ inches and weighed approximately.3 grams. The samples were 1% cotton lint collected from the lint screen during Task 3 tests. The table includes the weights of the lint samples tested, test conditions, and the observed results. Table 8. Testing on Top of the Heater Housing at Location B. Location Number Weight (grams) Power Intake Airflow (sfpm) Outside Airflow (sfpm) High-Limit Thermostat Result B s instant 4 bypassed ignited B s instant 4 bypassed ignited B s instant 4 bypassed ignited B s instant 4 series smolder B s instant 3 4 series smolder B s instant 2 4 series smolder In three of the six tests conducted at location B, the high-limit thermostat was bypassed; in the other three, the high-limit thermostat was connected in series with the heating element. The airflow through the heater was set at either 2 or 3 sfpm in the last two tests. In the first three tests at location B (s5-8, s5-9, and s5-1), the samples ignited. Figure 56 shows sample s5-8 igniting during the test. Figure 57 shows a graph of the thermocouple data for sample s5-8. The sample ignited approximately 3 seconds after power was applied to the heating element. 77

91 Sample 5-8 Before Smoldering Ignition Figure 56. Sample s5-8 at Location B Sample s grams July 12, T3 14 Temperature (C) Lint Sample Ignites T2 Power to Heater Removed Air Flow (sfpm) 2 T1 4 1 T Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Liint Sample T4 Exhaust Vent Heater T5 Exhaust Vent Outside Heater T6 Chamber T7 Room Ambient V1 Airflow into Heater V2 Airflow External Heater Figure 57. Thermocouple Data for Sample s5-8 In the last three tests, the high-limit thermostat was connected in series with the heating element. In the test for sample s5-11, the power was applied instantaneously to the heating element, and the sample smoldered until the high-limit thermostat opened and disconnected power to the heating element. For samples s5-12 and s5-13, the intake airflow into the heater 78

92 was 3 and 2 sfpm, respectively. The samples smoldered until the high-limit thermostat opened and disconnected power to the heating element Side of the Heater Housing Twenty-two tests were conducted with samples placed at three different locations (C, D, and E) on one side of the heater housing. All the samples were held in place with a finger probe. Three of the tests were conducted using lint samples obtained from a consumer s dryer and are labeled as hl in Tables 1 and 11 below. Table 9 provides a list of the lint ignition tests that were conducted at location C on the side of the heater housing. The high-limit thermostat was bypassed in all tests conducted at location C. The lint samples used measured approximately 2 x 2 x ¼ inches and weighed approximately.3 grams. The samples were 1% cotton lint taken from the lint screen collected during Task 3 testing. The table includes the weights of the lint samples tested, test conditions, and the observed results. Table 9. Testing on the Side of the Heater Housing at Location C Location Number Weight (grams) Power Intake Airflow (sfpm) Outside Airflow (sfpm) High-Limit Thermostat Result C s instant 65 4 bypassed C s instant 4 bypassed smolder C s instant 65 4 bypassed C s5-4.3 instant 4 bypassed smolder Four samples were tested at location C. Three of the samples (s5-2, s5-3 and s5-4) smoldered, and one (s2-7) ignited. Lint sample s2-7 was tested at location C, as shown in Figure 58. The T3 thermocouple was placed above the sample as an indicator of ignition. Figure 59 graphs the temperatures and airflow obtained during the test. Sample s2-7 was tested with 65 sfpm airflow (V1) initially through the heater box. At 4 seconds, the blower for the V1 airflow was stopped, and the temperature inside the heater rose rapidly, as expected. The sample ignited approximately 9 seconds after the blower was stopped. 79

93 T3 Thermocouple High-Limit Thermostat Sample 2-7 Finger Probe Figure 58. Sample s2-7 on the Side of the Heater Housing at Location C Sample 2-7 3s to 6s May 21, T2 Power to Heater Removed Temperature (C) V1 reduced to zero 4s Lint Sample Ignites Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Liint Sample T4 Exhaust Vent Heater T5 Exhaust Vent Outside Heater T6 Chmaber T7 Room Ambient V1 Airflow into Heater V2 Airflow External Heater T Air Flow (sfpm) Figure 59. Sample s2-7 Thermocouple Traces Table 1 provides a list of the lint ignition tests that were conducted at location D, which was adjacent to the high-limit thermostat. The table includes the weights of the lint samples tested, test conditions, and the observed results. 8

94 Table 1. Testing on Top of the Heater Housing at Location D Location Number Weight (grams) Power Intake Airflow (sfpm) Outside Airflow (sfpm) High-Limit Thermostat Result D s1-5.3 instant 43 bypassed ignited D s instant 65, 43 series D s2-1f.35 instant 43 bypassed smolder D s instant 43 bypassed smolder D hl1-1*.36 instant 43 bypassed smolder D s instant 4 bypassed smolder * Consumer lint Six lint samples were tested at Location D, as shown in Figure 6. The samples were held in place using a single finger probe. The T3 thermocouple was placed on the lint sample as an indicator of lint ignition. Four of the samples smoldered, one charred, and one ignited. Sample High-Limit Thermostat T3 Thermocouple Finger Probe Figure 6. Sample s2-2 Tested at Location D In five of the tests, including one using lint material from a consumer s home, the highlimit thermostat was bypassed, and there was no airflow through the heater box. Four of the tests resulted in smoldering of the lint sample (s2-1f, s2-2, hl1-1, and s5-5), and in one test, the sample ignited (s1-5). In a sixth test with the high limit thermostat in series with the heater, the lint sample charred (s2-1). The initial airflow was at 65 sfpm; at 4 seconds, the airflow was dropped to sfpm. The sample charred until the high-limit thermostat opened and disconnected power to the heating element. Table 11 below provides a list of the lint ignition tests that were conducted at location E, which was at a tab opening (punch-out) used to hold the heating element. Twelve tests were conducted at this location. The table includes the weights of the lint samples tested, test conditions, and the observed results. 81

95 Table 11. Testing on the Side of the Heater Housing at Location E Location Number Weight (grams) Power Intake Airflow (sfpm) Outside Airflow (sfpm) High-Limit Thermostat Result E s instant 43 bypassed ignited E s instant 8 4 bypassed charred E s1-7f.31 instant step 4 bypassed smolder down E s instant 43 bypassed ignited E hl1-2*.26 instant 43 bypassed ignited E s instant 43 bypassed ignited E hl1-3*.4 instant bypassed E s instant bypassed E s instant bypassed E s instant 43 bypassed ignited E s instant 4 bypassed ignited E s instant 4 bypassed ignited * Consumer lint In seven of the tests, including one using consumer lint material, the airflow into the heater was sfpm, and power to the heating element was applied instantaneously. All seven samples ignited (s1-6, s2-3, hl1-2, s2-4, s3-1, s5-6 and s5-7). Three tests were conducted with the airflow initially set at 65 sfpm; and at 4 seconds, airflow was dropped to sfpm. Two samples (s2-5, s2-6) ignited, and the consumer lint sample (hl1-3) smoldered. Figure 61 shows sample s2-6 before and after it ignited. Figure 62 shows the thermocouple traces for sample s2-6. The T3 thermocouple shows when the lint sample ignited by the rapid increase in temperature. The sample ignited approximately 6 seconds after the blower was turned off. Sample 2-6 Before Sample2-6 Ignition Figure 61. Sample s2-6 Tested at Location E 82

96 Sample grams, Location E May 21, Lint Sample Ignites T2 Power to Heater Removed Temperature (C) V1 reduced to zero 4s T Air Flow (sfpm) Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Liint Sample T4 Exhaust Vent Heater T5 Exhaust Vent Outside Heater T6 Chmaber T7 Room Ambient V1 Airflow into Heater V2 Airflow External Heater Figure 62. Thermocouple Traces for Sample s2-6 at Location E A single test was conducted with the airflow set at 8 sfpm. The sample (s1-7) charred slightly. The last test was conducted with the voltage to the heating element stepped down incrementally; the sample (s1-7f) smoldered (Appendix F shows the temperature and airflow data for sample s1-7f) Lint Samples at the Heater Intake Lint samples were placed at various distances from the heater intake, as shown in Figure 63. The samples were approximately 1 x 1 x 1/4 and weighed approximately.2 grams. The samples were held in place by a finger probe. The T3 thermocouple was placed above the sample as an indicator of ignition (Appendix G illustrates the setup). The lint samples were placed at 1-inch increments from the front edge of the heater housing. The distance from the heating element to the edge of the heater housing was 2 inches. 83

97 T3 Thermocouple Lint Sample Finger Probe Figure 63. Lint Sample at Heater Intake An infrared camera was positioned to view the lint samples, as shown in Figure 64, and was used to capture the temperature of the lint during the test. Infrared data was not captured for all of the lint samples tested. IR Camera Cooling Fan Heater Box Video Figure 64. Infrared Camera Setup 84

98 For the following test series, the airflow outside the heater boxes for the different dryer designs measured between and 43 sfpm depending on whether the exhaust vent was blocked or unblocked. The airflow over the heater box/housing was set at approximately 43 sfpm to simulate a slight airflow caused by the drum rotating and intake into the heater. The first series of tests was conducted with no airflow through the heater box and the high-limit thermostat bypassed. In this series of tests, the lint samples ignited at a distance ranging from 3 to 4 inches from the housing edge. A second series of tests was conducted similar to the first set, except that the high-limit thermostat was connected in series with the heating element. The samples ignited at a distance between 1 to 2 inches from the housing edge. A third series of tests was conducted similar to the second series but with airflow through the heater box. The lint samples were fixed at a distance of one inch from the heater edge. The samples ignited when the airflow at the heater intake was between 3 and 4 sfpm. A single test was conducted with a lint sample placed 1-inch from the edge of the heater housing, the high-limit thermostat bypassed, and the airflow through the heater box set at 4 sfpm. The lint sample ignited. Table 12 lists the tests conducted with lint samples at the heater intake. The list is presented in chronological order (the order in which the tests were conducted). The distance is the number of inches from the heater housing edge to the front face of the sample. The column labeled High-Limit Thermostat indicates whether the thermostat was connected in series with the heating element or not ( yes means the thermostat was connected in series; no means the thermostat was bypassed.) 85

99 Table 12. Lint Samples at the Heater Intake Lint Weight Distance High-Limit Airflow Result Sample # (grams) (inches) Thermostat (sfpm) s no ignition s no ignited s no no ignition s no ignited s no ignited s no ignited s no ignited s no ignited s no ignited s no no ignition s no no ignition s3-13*.19 2 yes ignited s yes no ignition s yes ignited s yes ignited s yes ignited s yes no ignition s yes 2 ignited s yes 4 no ignition s yes 3 ignited s1-9.2 yes 4 no ignition s no 4 ignited * The high-limit failed; s3-13 is later discussed in the Discussion Section, Task 4 Figure 65 shows a sequence of images from the infrared camera and a video camera for Sample s3-1. The figure shows the lint sample began to ignite at approximately 452 C. Figure 66 shows the thermocouple data for sample s3-1. The sample ignited 1 minute 11 seconds into the test. The sample ignited approximately 4 seconds after the heater was energized. 86

100 Smoldering :5 Sample 3-1 Sample 3-1 Ignition 1:11 Sample 3-1 Sample 3-1 Flames 1:13 Sample 3-1 Sample 3-1 Figure 65. Sample s3-1, Video Camera and Infrared Camera 87

101 Sample 3-1 weight.16g sample 3" from edge June 4, T3 Power to Heater turned off Temperature (C) Ignition T2 T Air Flow (sfpm) 1 T Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Lint Sample T4 Exhaust Vent Heater T5 Exhaust Vent Outside Heater T6 Chamber T7 Room Ambient V1 Airflow into Heater V2 Airflow External Heater Figure 66. Sample s3-1 Thermocouple and Airflow Data 88

102 2.4.4 Phase II Ignition Characteristics of Lint Ingested into the Heater Box This phase of testing was divided into two parts. Part 1 of the tests involved observation of the heater exhaust as the lint samples of varying weights and sizes were ingested into the heater box. Part 2 was to observe whether target materials downstream of the heater exhaust could be ignited by embers as they exited the heater exhaust. The test setup was designed to allow the airflow into the heater intake to carry lint samples past the heating element and exhaust any embers through the ducts. To observe the heater exhaust, a 28-inch section of the metal duct from Phase I tests was replaced with a 4 diameter high-temperature glass tube, as shown in Figure 67. (The actual dimensions for the test setup can be found in Appendix H.) Heater Glass Tube View Ports Airflow Direction Figure 67. Setup with Glass Tube Part 1 Lint Samples Ingested by the Heater Box For these tests, different weights and sizes of lint were drawn into the heater box. Six series of tests were completed: 1. The first series included 5 samples weighing between.24g and.77g. 2. The second series consisted of 8 samples weighing between.4g and.18g. 3. The third series consisted of 5 samples weighing between.14g and.48g. 4. The fourth series consisted of 1 samples at.1 ±.1 grams. 5. The fifth series consisted of 1 samples at.2 ±.1 grams. 6. The sixth series consisted of 1 samples at.3 ±.1 grams. 89

103 Table 13 below lists the six series of tests and the lint sample weights used in each. Table 13. Lint Samples for Task 4 - Phase II Testing (Observation of Samples Ingested into Heater Box) Series Number Sample Number Weight (grams) to to to 1.3 For all tests conducted, embers were visible exiting the heater box. The embers varied in size and number. It was observed that larger lint samples produced more and/or larger embers than smaller lint samples did, but two lint samples of the same size did not produce the same ember size and/or number of embers. The number and size of the embers in the heater exhaust were dependent on how and where the lint samples contacted the heating element. For example, if the lint sample became trapped on the heating element, smaller embers were produced in the exhaust as the sample burned. However, if the sample contacted the heating element, ignited, and continued through the heater box, larger embers were observed exiting the heater exhaust Part 2 Ignition of Target Materials Downstream from the Heater Exhaust Four series of tests were conducted to observe whether target materials downstream of the heater exhaust could be ignited by embers as they exited the heater exhaust. Target materials were placed inside the glass tube, as shown in Figure 68. After the target material was placed in the glass tube, lint samples were drawn into the heater box approximately every 3 9

104 to 4 seconds. The airflow was set at 8 sfpm at the heater intake. This airflow velocity simulated the airflow inside a dryer near the end of the drying cycle (when the dryer would have a full lint screen). For each of the four tests, the lint samples drawn into the heater box consisted of ten samples of.1g each, then 1 samples of.2g each and, finally, ten samples of.3g. Glass Tube Target Material Airflow Direction Test 1 Figure 68. Setup with Target Material in the Glass Tube Clips to secure material Table 14 outlines the types, sizes and weights of the target materials used for each test series. The target samples were either sheets of 1% cotton lint or cotton terry towel material. The lint target samples weighed between 2 and 5 grams and measured approximately 6 x 1 inches. The lint target material varied in thickness from 1/8 to ½ inch. The lint target materials were collected during Task 3 testing. The cotton terry towel material was a section cut from a towel used in Task 3 testing; the section measured 7 ½ x 1 inches and weighed 35.2 grams. Table 14. Target Materials for Task 4 - Phase II Testing (Ignition of Target Materials by Exhaust Embers) Test Number Target Material Size (inches) Weight (grams) 1 Lint Sheet 1 x 6 ½ x ½ to Lint Sheet 7 x 5 ½ x ½ x Lint Sheet 7 ½ x 6 ¼ to Cotton Towel 7 ½ x

105 In all four tests, the target material ignited: 1. In Test 1, the target material ignited on the fifth lint sample of.1 g. 2. In Test 2, the target material ignited on the sixth sample of.2 g. 3. In Test 3, the target sample ignited on the tenth sample of.1 g. 4. In Test 4, the target material ignited on the eighth sample of.3 g. Figure 69 shows the target sample igniting during Test 1. The sample was consumed in less than 6 seconds and produced additional embers downstream. The exhaust temperature, T4 thermocouple, increased to over 26 C in less than 5 seconds, as shown in Figure 7. (Test data for Tests 2, 3, and 4 are contained in Appendix H. Video clips of the lint sample testing are included in Appendix J.) 92

106 -:1 Embers :2 : Ignition :3 :1 :4 Figure 69. Target Sample Igniting During Test 1 93

107 Test 1 Target Material - Lint - weight 5 grams June 14, Heater Turned Off Temperature (C) Ignition T Air Flow (sfpm) Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Glass Tube Internal T4 Exhaust Vent Heater T5 Exhaust Vent Outside Heater T6 Chamber T7 Room Ambient V1 Airflow into Heater V2 Airflow External Heater Figure 7. Test 1 Thermocouple and Airflow Measurements Task 4- Test 1 (Ignition of Target Materials by Exhaust Embers) 94

108 3. DISCUSSION This section of the report presents a discussion of the tests and the test results reported in Section 2. Description of Test and Test Results for each of the four project tasks. Each discussion section is supported by its corresponding section in Section 2., unless otherwise specified. 3.1 Task 1: Inspect and Record Dryer Design Four clothes dryers (Dryer Designs A, B, C and D) were used for testing. As part of each dryer s exhaust system, there was a 4-inch diameter male duct, a short length of which protruded outside the dryer housing. This short length of duct was slightly tapered and was intended for connection to external venting (the external vent slid over the tapered male duct). The external vent material used for the clothes dryers was 4-inch diameter rigid metal ducting, as recommended in the manufacturers installation instructions. The interface of the male duct and the external venting was sealed and/or secured with a 4-inch hose clamp. The hose clamp nut was tightened by hand with a nut driver until it could not be tightened any further. The amount of torque applied to the hose clamp nut had little effect on actually tightening the connection between the male duct and the external vent. Two factors contributed to the poor seal: 1. The length of the dryer s male duct (which varied slightly among the designs) was insufficient to allow the external vent to slide far enough onto the duct to provide a secure pressure fit. 2. Both mating pieces were rigid; thus, the external venting could not compress around the dryer s male duct when the hose clamp was tightened. (Additional comments regarding dryer designs are included in the Discussion section of Task 3: Monitor Lint Distribution.) 3.2 Task 2: Document Dryer Operating Characteristics Testing showed that each dryer design reacts slightly differently when the exhaust vent is unblocked, partially blocked, or 1% blocked. The data for temperature and airflow varied slightly with each dryer, but certain characteristics were similar for all the dryer designs tested Normal Operation (Unblocked Exhaust Vent) When the dryers were tested with an unblocked exhaust vent, similar temperatures the intake air temperature into the heater, the heater housing temperature, the intake air temperature into the blower, and the exhaust vent temperature were measured among all dryer designs. Figures 71 through 75 show comparisons of temperatures measured for each of the dryer designs when tested with a wet load. Dryer Design D used only 7 towels so that the dryer would not operate in the high-limit cycling mode. (Additional graphs demonstrating dryer design comparisons can be found in Appendix I). 95

109 Figure 71 shows that the heater intake temperature (T1) was higher for Dryer Design C than for other dryer designs, and it reached a maximum of 75 C. Dryer Designs A and C exhibited similar maximum temperatures before the primary thermostat opened. Dryer Design B had a slightly lower maximum temperature. Since the load in Dryer Design D began riding the drum after approximately 19 seconds, the airflow through the tumbler was less restricted causing the primary thermostat to reach its opening set point sooner. Figure 72 shows the heater housing temperatures did not extend much higher than 12 C for all the dryer designs tested. The housing temperatures for Dryer Designs A and D were similar until the load in Dryer Design D stopped tumbling after approximately 19 seconds. The heater housing in Dryer Design B had a significant number of punch-outs on one side, which were used to mount the heating element inside the heater housing. This appeared to allow the heater housing to operate at a slightly cooler temperature than either Dryer Design A or D. Dryer Design C had louvers on one side of the heater housing; this allowed the heater housing to operate at an even lower temperature approximately 66 C, as shown in the graph. Figure 73 shows the intake temperature into the blower for all the dryer designs tested. The temperatures are very similar for all the dryers, as shown in the graph. The intake temperature into the blower did not extend much higher than 1 C for all the dryer designs tested. Dryer Design A and D had a slightly higher operating curve than the other dryers. The load in Dryer Design D began to stop tumbling at approximately 19 seconds after the dryer was started, as shown in the graph. Figure 74 shows the exhaust vent temperatures for all the dryer designs tested. The temperatures are very similar for all the dryers, as shown in the graph. The exhaust venting temperature did not extend much higher than 7 C for all the dryer designs tested. The exhaust vent temperatures for Dryer Design A were slightly higher than those for the other dryers. Figure 75 shows the maximum temperatures measured at each thermocouple location for each dryer design. All the dryers had similar temperatures at each thermocouple location, except at the heater exhaust and tumbler intake. Design characteristics had a contributing factor on temperatures at these locations. Dryer Design C had an increase in temperature from the heater exhaust to the tumbler intake, which was also observed during testing in Task 2. The louvers on the side of the heater housing may have contributed to the higher temperature readings at T6, but it has not yet been determined why this actually occurred. 96

110 T1 Heater Intake Temperature Dryer Design Comparison Wet Load A C Temperature (C) D B Time (seconds) Design A Design B Design C Design D Figure 71. Dryer Design Comparison for T1, Heater Intake T3 Heater Housing Temperature Dryer Design Comparison Wet Load D Temperature (C) A B C Time (seconds) Design A Design B Design C Design D Figure 72. Dryer Design Comparison for T3, Heater Housing 97

111 T5 Intake into Blower Temperature Dryer Design Comparison Wet Load Dryer Design D Load Stops Tumbling B Temperature (C) A D C Time (seconds) Design A Design B Design C Design D Figure 73. Dryer Design Comparison for T5, Blower Intake T4 Exhaust Vent Temperature Dryer Design Comparison Wet Load 8 7 B Temperature (C) A D C Time (seconds) Design A Design B Design C Design D Figure 74. Dryer Design Comparison for T4, Exhaust Vent 98

112 Dryer Design Comparison Maximum Temperature at Test Locations Wet Load Temperature (C) A B C D Analog Signal Not Temperature 5 Heater Intake Heater Exhaust Heater Housing Dryer Vent Intake to Blower Intake into the Tumbler Room Temperature Location Design A Design B Design C Design D Figure 75. Maximum Temperatures Measured at Location Partially-Blocked and 1%-Blocked Conditions Each dryer design reacted slightly differently when the exhaust vent was partially blocked and/or 1% blocked. The responses for each dryer design were discussed in Task 2 of the Testing section. The discussion below examines the general characteristics of a dryer by analyzing the data without compensating for variances in dryer designs. The data presented in this section are in general terms and do not represent a specific dryer tested. The data was generated from Dryer Design A/dry load, Dryer Design B/wet load, Dryer Design C/wet load, and Dryer Design D/no load. Dryer Design A/dry load was used since the operating temperatures for a dry load were similar to a wet load, although the time scale was different. Dryer Design D/no load was used to avoid using data in which the dryer quickly began operating in the high-limit cycle mode for a full load or in which operation was unpredictable (the load stopped tumbling). The maximum temperatures measured for each dryer for each test condition (blocked exhaust vent, partially blocked vent, or unblocked vent) were extracted from each data set. The temperatures were then categorized as highest, lowest, and average for each set. Figures 76 through 81 show comparisons of maximum temperatures measured at each test location under unblocked conditions, the range of blocked conditions, and with the primary thermostat bypassed. 99

113 Figure 76 shows the range of highest to lowest maximum intake air temperatures at the heater for each test condition. The temperatures did not change significantly up to a 5% blockage of the exhaust vent. At 75% blockage, the temperatures (highest, average, and lowest) increased significantly, with the highest temperature measuring over 35 C. The average maximum temperature was slightly over 15 C with a 75% blockage. The maximum temperatures when the exhaust vent was 1% blocked were similar to the 75% blocked condition. The maximum temperatures with the primary thermostat bypassed were similar to the temperatures when the exhaust vent was 5% blocked. Figure 77 shows the range of highest to lowest maximum heater exhaust temperatures for each test condition. The temperatures did not change significantly from the unblocked exhaust vent condition to a 25% blocked exhaust vent. As the exhaust vent was blocked beyond 25%, there was a slow increase in temperatures. The highest maximum temperature measured nearly 6 C for the 1% blocked condition. The maximum temperatures with the primary thermostat bypassed were similar to the temperatures when the exhaust vent was 75% blocked. Figure 78 shows the range of highest to lowest maximum heater housing temperatures for each test condition. The temperatures generally showed a slight increase from the unblocked exhaust vent condition to a 5% blocked exhaust vent. When the vent was 75% blocked, the maximum temperatures increased more rapidly. When the exhaust vent was 1% blocked, the maximum temperatures were similar to the temperatures reached for a 75% blocked vent. The highest maximum temperature measured between 2 C and 25 C for the 75% and 1% blocked conditions. The maximum temperatures with the primary thermostat bypassed were similar to those for a 75% blocked exhaust vent. Figure 79 shows the range of highest to lowest maximum dryer exhaust vent temperatures for each test condition. The temperatures did not change significantly from the unblocked exhaust vent condition to a 5% blocked exhaust vent. Beyond a 5% blockage, the maximum temperatures measured began to drop rapidly. The highest maximum temperature measured dropped from approximately 7 C to 3 C. The maximum temperatures measured when the primary thermostat was bypassed were much higher than those measured for any of the blocked or unblocked exhaust vent conditions. The highest maximum temperature measured was slightly over 1 C. Figure 8 shows the range of highest to lowest maximum temperatures at the intake into the blower for each test condition. The temperature curves shown in this graph are similar to those for the dryer exhaust vent (Figure 79), although the changes were not as dramatic. Another similarity between Figures 8 and 79 is that, when the primary thermostat was bypassed, maximum temperatures were much higher than those measured for any of the blocked or unblocked exhaust vent conditions. The highest maximum temperature measured was slightly over 12 C. Figure 81 shows the range of highest to lowest maximum temperatures at the intake into the tumbler for each test condition. The temperature curves shown in this graph are similar to those for the heater exhaust (Figure 77), except that the maximum temperature measured when the exhaust vent was fully blocked was lower approximately 4 C. Another similarity between Figures 81 and 77 is that, when the primary thermostat was bypassed, maximum temperatures were similar to those for the 75% blocked condition. 1

114 All Dryers Intake into the Heater Temperature C maximum average minimum % 25% 5% 75% 1% Bypassed Percent Blocked Figure 76. Intake into the Heater All Dryers Heater Exhaust Temperature C maximum average minimum % 25% 5% 75% 1% Bypassed Percent Blocked Figure 77. Heater Exhaust Temperature 11

115 All Dryers Heater Housing Temperature 25 2 Temperature C maximum average minimum % 25% 5% 75% 1% Bypassed Percent Blocked Figure 78. Heater Housing Temperature All Dryers Exhaust Vent Temperature 12 1 Temperature C maximum average minimum 2 % 25% 5% 75% 1% Bypassed Percent Blocked Figure 79. Exhaust Vent Temperature 12

116 All Dryers Intake into the Blower Temperature C maximum average minimum % 25% 5% 75% 1% Bypassed Percent Blocked Figure 8. Intake into the Blower Temperature All Dryers Heated Air into the Tumbler Temperature C maximum average minimum % 25% 5% 75% 1% Bypassed Percent Blocked Figure 81. Intake into the Tumbler Temperature 13

117 Figure 82 shows the data from Figures 76 through 81 in an illustrated dryer configuration. The dryers main components are illustrated in a linear representation. The figure shows the maximum temperatures for all the dryer designs at each location for the range of unblocked and blocked exhaust vent conditions, as well as for when the primary thermostat was bypassed. The figure shows that there was a significant increase in maximum temperatures at the intake into the tumbler when the vent was 75% and 1% blocked. The air temperatures exiting the tumbler and entering the exhaust vent decreased, as expected. For the unblocked, 25% blocked and 5% blocked exhaust vent conditions, the maximum temperatures did not show a significant change. When the primary thermostat was bypassed, there was little change in maximum temperatures at T1 (heater intake) and T3 (heater housing). There was an increase in the maximum temperatures after the heater exhaust and into the exhaust vent. 14

118 Heater Element Tumbler Lint Screen Blower Intake Air Exhaust Air T1 Heater Intake T3Heater Housing T2 Heater Exhaust T6 Tumbler Intake T5 Blower Intake T4 Exhaust Vent Figure 82. Maximum Temperatures for All Dryers at Each Location and Condition 15

119 Tables 15 through 17 show the percentage of change in minimum, average and maximum temperatures for the different percentages of exhaust vent blockage conditions (and a bypassed primary thermostat) as compared to the unblocked condition for all dryer designs. There was an increase in temperature from the heater intake to the tumbler for the partially and fully blocked conditions. As shown in Table 17, there was as increase in temperature ranging from 69% to 351% from the unblocked to 1% blocked conditions, depending on the location. After the tumbler, the temperature increased between 13% to 55% from the unblocked to 1% blocked conditions, depending on the location, as expected. When the primary thermostat was bypassed, the temperature after the tumbler was higher by 38% to 48% when compared to the unblocked condition. Percent Blocked Table 15. Change in Minimum Temperature Percent Change in Minimum Temperature from Unblocked Condition T1 Heater Intake T3 Heater Housing T2 Heater Exhaust T6 Tumbler Intake T5 Blower Intake T4 Exhaust Vent 25% 1% % 2% -1% -2% -2% 5% -4% 2% 32% 37% -2% -3% 75% 34% 29% 56% 53% -21% -2% 1% 39% 41% 3% 26% -37% -56% Bypassed 1% -1% 24% 41% 41% 35% Percent Blocked Table 16. Change in Average Temperature Percent Change in Average Temperature from Unblocked Condition T1 Heater Intake T3 Heater Housing T2 Heater Exhaust T6 Tumbler Intake T5 Blower Intake T4 Exhaust Vent 25% % 1% % -1% % -1% 5% 3% 11% 8% 7% 4% 1% 75% 137% 81% 37% 24% -2% -9% 1% 162% 1% 78% 39% -3% -55% Bypassed 39% 2% 21% 18% 34% 43% Percent Blocked Table 17. Change in Maximum Temperature Percent Change in Maximum Temperature from Unblocked Condition T1 Heater Intake T3 Heater Housing T2 Heater Exhaust T6 Tumbler Intake T5 Blower Intake T4 Exhaust Vent 25% 1% -7% -1% -1% % 1% 5% 6% 6% 3% 3% % 5% 75% 386% 93% 33% 23% 6% -1% 1% 351% 92% 89% 69% -13% -55% Bypassed 43% 15% 23% 17% 36% 48% 16

120 Figure 83 shows the exhaust air velocities with an unblocked vent for the different dryer designs. The graph shows that the air velocities for Dryer Designs A and D were similar throughout the 6-minute drying cycles. The graph shows that the air velocities for Dryer Designs B and C were also similar. In addition, Dryer Designs B and C showed a steady decline in exhaust air velocities as the drying cycle progressed. Velocity Exhaust Comparison All Dryer Designs Wet Load Velocity (sfpm) Time (seconds) Design A Design B Design C Design D Figure 83. Comparison of Exhaust Air Velocity for all Dryer Designs Figure 84 shows the average exhaust air velocity with an unblocked vent for all the dryer designs. The graph shows a steady decline in the exhaust velocity during the drying cycle. The average velocity drops from 13 sfpm to 9 sfpm. Average Velocity Exhaust Wet Load 2 18 Velocity (sfpm) Time (seconds) Average Figure 84. Calculated Average Exhaust Air Velocity 17

121 3.3 Task 3: Monitor Lint Distribution Based on the test data and observations presented in Section 2.3, even when the lint filter was cleaned after each use of the dryer, there was visible lint accumulation inside the chassis. This section further evaluates the mechanisms within dryers that may allow lint to leak into the dryer structure. The areas of positive pressure in the airflow were the most likely causes for lint leaking into the dryer chassis Dryer Design A at Positive Pressure Examination of Dryer Design A showed that lint/air can leak through a sealed joint in the blower housing. To examine more closely the nature of the origin of the lint/air leak, the blower fan compartment was opened on a blower from another dryer of Design A. Figures 85(a) and (b) show the blower housing assembled and opened, respectively. (a) Blower removed from the dryer (b) Blower housing separated Figures 85 (a) and (b). Dryer Design A Blower Assembly The blower is a centrifugal type fan. The blower operates by sucking air through the center hole at negative pressure and forcing the air outward at positive pressure as shown in 18

122 Figure 86. It is noteworthy that a foam gasket was used to seal the outer portion where there is positive pressure. HEATER DRUM LINT SCR FAN DRYER CHASSIS DRYER CHASSIS EXTERIOR ATMOSPHERIC PRESSURE BELOW ATMOSPHERIC PRESSURE ABOVE ATMOSPHERIC PRESSURE Figure 86. An Illustration of Airflow through the Dryer and Blower Other Dryer Designs at Positive Pressure The other dryer designs (B, C, and D) did not contain the same configuration that resulted in positive pressure after the blower. As shown in Dryer Design A, the blower was at the rear of the dryer chassis. This configuration allowed the blower housing and connection to the external dryer vent to be the same component. In the other dryer designs, the blower was located toward the front of the dryer chassis, where the lint screen was also located. These dryers required a short section of duct between the rear of the dryer and the blower housing. The section of the dryer vent/ducting after the blower and inside the chassis was an area of positive pressure. Most importantly, the connection between the short section of duct and the blower was at positive pressure. Figure 87 shows the junction was at positive pressure (white arrow). Though the fitting of the metal duct to the plastic nozzle of the fan was snug, it was not airtight. In some case, there was no seal between the blower housing and the short section of exhaust vent. 19

123 Dryer Design B Figure 87. Junction Between the Blower and Exhaust Vent Under the test conditions, which included rigid exhaust ducting, two 9 elbows, and no blockage in the ducting, there was lint accumulation in the chassis interior. The setup produced minimal backpressure in the dryer exhaust vent (See Appendix E). In the event that a serious blockage did occur, lint leakage could be expected to increase as the backpressure in the exhaust vent increased. The extent to which leakage would be made worse would depend on the location of the leakage. Close to the blower impeller on the side opposite of the exhaust port yet inside impeller housing the positive pressure is greatest. 11

124 3.4 Task 4: Determine Characteristics Required for Lint Ignition Based on the test data and observations presented in Section 2.4, ignition characteristics of lint were observed on and near the heater housing (heater box) and when lint was ingested into the heater box. This section further assesses the ignition times and temperatures under the various conditions tested. This section also includes an examination of high limit devices that had failed during the course of testing Lint on the Heater Housing Twenty-two lint samples were tested at five different locations on the heater housing. The samples at either locations B or E were more likely to ignite than samples placed at locations A, C, or D on the heater housing. The samples located at B and E ignited more frequently under conditions of no airflow and when the high-limit thermostat was bypassed. The samples would smolder but generally not ignite when placed at the other locations and/or when tested under other conditions. Instead, the lint material would be consumed by pyrolysis or thermal decomposition Lint Samples at the Heater Intake This phase of testing was divided into three test series. The first series of tests was conducted under conditions of no airflow through the heater box, and the high-limit thermostat bypassed. This simulated a condition in which the exhaust vent was blocked and the high-limit thermostat had failed to open. The second test series was conducted with no airflow through the heater box, but the high-limit thermostat was connected in series with the heating element. This simulated a condition where the exhaust vent was fully blocked, but the high-limit thermostat was operational. The third series of tests was conducted with various airflow velocities through the heater box, and the high-limit thermostat connected in series with the heating element. This simulated a partially blocked exhaust vent with an operational high-limit thermostat. In these tests, the primary purpose of the T3 thermocouple, which was placed on the lint sample, was to indicate when the sample ignited. The T3 thermocouple temperature data should not be used as an analysis tool to compare samples, since the placement of the thermocouple on the samples may not have been identical. The thermocouple was placed above the sample, to achieve the maximum change in temperature should the sample ignite High-Limit Thermostat Bypassed and No Airflow Eleven tests were conducted in the first series of tests. Lint samples were placed at distances ranging from 2 to 4 inches from the edge of the heater box. (This would equate to 4 to 6 inches from the heater element.) The data shows the threshold for ignition and no ignition was approximately 3 inches and 4 inches (from the edge of the heater housing), respectively. (Appendix I shows data for the lint sample tested 2 inches from the heater intake.) Figure 88 below shows the temperatures at the T1 Heater Intake, T2 Heater Exhaust and T3 Lint Sample thermocouples for samples s3-3, s3-9, and s3-1, which were tested 3 inches from the heater housing edge. (The heating element was energized approximately 35 seconds into the test.) 111

125 Samples s3-9 and s3-1 show similar curves for the T1 thermocouple as shown in Figure 89. For tests with sample s3-3, the T1 thermocouple showed slightly lower temperatures, which coincides with the later ignition time shown by T3 thermocouple. The temperature curves for the T2 thermocouple were similar for all three samples tested as shown in Figure 9. The times to ignition are similar for samples s3-9 and s3-1 approximately 83 and 81 seconds after powering the heating element, respectively. Sample s3-3 had a slightly longer time to ignition approximately 17 seconds after powering the heating element as shown in Figure 91. Lint Sample Intake at 3 inches High Limit Bypassed, No Air flow 22 7 Temperature (C) s3-9 T1 s3-3 T1 s3-1 T2 s3-1 T3 s3-9 T3 s3-9 T2 s3-3 T2 s3-3 T s3-1 T1 s3-3 T1 s3-3 T2 s3-3 T3 s3-9 T1 s3-9 T2 s3-9 T3 s3-1 T1 s3-1 T2 s3-1 T3 Time (seconds) Figure 88. Lint Samples Tested 3 Inches from the Heater Housing Edge 112

126 T1 Only - Lint Sample Intake at 3 inches High Limit Bypassed, No Airflow Temperature (C) T1 3-9 T1 3-1 T1 average T Time (seconds) Figure 89. T1 Only - 3 Inches from the Heater Housing Edge T2 Only - Lint Sample Intake at 3 inches High Limit Bypassed, No Airflow Temperature (C) T2 3-9 T2 3-1 T2 average T2 Time (seconds) Figure 9. T2 Only - 3 Inches from the Heater Housing Edge 113

127 T3 Only - Lint Sample Intake at 3 inches High Limit Bypassed, No Airflow Temperature (C) T3 3-9 T3 3-1 T3 average T Time (seconds) Figure 91. T3 Only - 3 Inches from the Heater Housing Edge Figure 92 shows the temperatures at the T1, T2 and T3 thermocouples for samples s3-4, s3-11, and s3-12, which were tested 4 inches from the heater housing edge. (The heating element was energized approximately 35 seconds into the test.) All three samples smoldered, but no flames were present. The T1, T2 and T3 thermocouple traces were similar for all three samples tested. 114

128 Lint Sample Intake at 4 inches High Limit Bypassed, No Airflow Temperature (C) s3-12 T2 s3-11 T2 s3-4 T2 s3-4 T1 s3-11 T1 s3-12 T3 s3-12 T s3-11 T3 s3-4 T1 s3-4 T2 s3-4 T3 s3-11 T1 s3-11 T2 s3-11 T3 s3-12 T1 s3-12 T2 s3-12 T3 s3-4 T3 Time (seconds) Figure 92. Lint Samples Tested 4 Inches from the Heater Housing Edge Table 18 lists the ignition times for samples tested at 2 and 3 inches from the heater intake edge. Figure 93 plots the minimum, maximum, and average ignition times for samples tested at 2 and 3 inches from the heater edge. Sample Number Table 18. Ignition Times at Heater Intake 3 inches from Heater Intake 2 inches from Heater Intake Data Heater Actual Sample Data Heater Time Energized Time Number Time Energized (sec) (seconds) (sec) (sec) (seconds) Actual Time (sec) s s s s s s max 17 s average 9 s min 81 max 73 average 47 min

129 Ignition Times for Lint Samples 2 and 3 inches from Intake 4 Sample Distance Time (seconds) Figure 93. Ignition Time for 2 and 3 Inches from Heater Edge High-Limit Connected in Series and No Airflow Six tests were conducted in this series of tests in which the high-limit thermostat was connected in series with the heating element and there was no airflow through the heater box. Lint samples were placed at distances ranging from to 2 inches from the edge of the heater box (2 to 4 inches from the heating element, respectively). The data show a threshold of between 1 and 2 inches from the heater housing edge for ignition and no ignition, respectively. In the first test, sample s3-13 ignited, and the high-limit thermostat did not open and disconnect power to the heating element. It was determined that the high-limit thermostat had failed. (An investigation of the cause of failure for the high-limit thermostat was conducted, and the results are reported at the end of this Discussion Section.) The remainder of the discussion in this subsection, therefore, will not include data from sample s3-13. The high-limit thermostat was replaced with a similar high-limit thermostat taken from a dryer of the same make and model. Several baseline tests were conducted to confirm that the replacement thermostat operated properly when the temperature inside the heater housing reached a set temperature. Figure 94 shows the T1, T2, and T3 thermocouple data for samples s3-16 and s3-17, tested at a distance of 1 inch from the heater edge. The times to ignition for samples s3-16 and s3-17 were approximately 55 and 65 seconds after powering the dryer, respectively. The steep increases observed in the data for thermocouple T3 indicate ignition of the sample. The profiles of the data obtained from thermocouples T1 and T2 were similar for both samples tested. This figure shows that the samples ignited before the high-limit thermostat opened and disconnected power to the heating element. For both samples, the sample ignited 116

130 approximately 5 to 8 seconds before the high-limit thermostat opened, which is shown by the sudden decrease in temperature for the T1 thermocouple. 5 Lint Sample Intake 1 inch High Limit, No Airflow 22 Temperature (C) s3-16 T3 High-Lmit Opens s s3-17 T3 99 s3-17 T1 Time (seconds) s3-17 T2 s3-16 T s3-16 T1 s3-16 T2 s3-16 T3 s3-17 T1 s3-17 T2 s3-17 T3 Figure 94. Lint Samples s3-16 and s3-17 Tested 1 Inch from the Heater Edge Figure 95 shows the T1, T2, and T3 thermocouple data for samples s3-14 and s3-18, tested at a distance of 2 inches from the heater edge. Both samples smoldered, and no flames were present before the high-limit thermostat opened and disconnected power to the heating element. The profiles of the data obtained from thermocouples T1, T2 and T3 were similar for both samples tested. 117

131 3 Lint Sample Intake at 2 inches High Limit, No Air flow s3-18 T2 Temperature (C) s s3-14 T s3-14 T2 s3-18 T3 99 Time (seconds) s3-14 T s3-14 T1 s3-14 T2 s3-14 T3 s3-18 T1 s3-18 T2 s3-18 T3 Figure 95. Lint Samples s3-14 and s3-18 Tested 2 Inches from the Heater Edge In previous tests in which the high-limit thermostat was bypassed, a lint sample placed 2 inches from the heater edge ignited. Table 19 lists the high-limit thermostat activation times. The average activation time was 39 seconds. Figure 96 shows a similar figure as in Figure 93 but with the average high-limit thermostat activation time at 93 seconds. The figure indicates the high-limit thermostat opens 8 seconds before the average sample at 2 inches from the heater edge would ignite. In some cases, a lint sample 2 inches from the heater edge may ignite before the high limit activates as shown in the figure. Table 19. Approximate High Limit Thermostat Activation Times Sample Number High Limit Activation 2 inches from Heater Intake with High Limit Data Heater Time Energized (seconds) (seconds) Actual Time (seconds) s s average

132 Ignition Times for Lint Samples 2 and 3 inches from Intake 4 Sample Distance 3 2 High Limit Activation Time (seconds) Figure 96. High Limit Activation Time Comparison at Sample 2 and 3 inches High-Limit Connected in Series and Airflow Four tests were conducted in this last series of tests, in which the high-limit thermostat was connected in series with the heating element and airflow through the heater box was varied. Lint samples were placed at distances either or 1 inch from the edge of the heater box (either 2 or 3 inches from the heating element). Airflow velocity, as measured at the intake, was varied from 2 to 4 sfpm. For a sample placed 1 inch from the heater housing edge, the thermocouple and anemometer data showed a threshold between 3 and 4 sfpm for ignition and no ignition, respectively. Figure 97 shows T1, T2, and T3 thermocouple data for samples s3-19, s3-2, and s1-8, which were tested at a distance of 1 inch from the heater edge with different airflow velocities. The graph shows that samples tested with airflow below 3 sfpm ignited. With an airflow of 2 sfpm, the sample ignited approximately 6 seconds after powering the heating element. With an airflow of 3 sfpm, the sample ignited approximately 8 seconds after powering the heating element. The graph also shows that samples tested with airflow below 3 sfpm ignited before the high-limit thermostat opened and disconnected power to the heating element. For both samples tested at 2 and 3 sfpm airflow, the samples ignited approximately 1 seconds before the high-limit thermostat opened (as shown by the decrease in temperature measured at thermocouple T2). Since thermocouple T1 was in close proximity to the lint sample, the thermocouple data for T1 and T3 showed similar responses. T2 thermocouple was used to approximate the time at which the high-limit thermostat opened and disconnected power to the heating element. Sample s1-9 was placed 1 inch from the heater housing edge; the airflow was set at 4 sfpm, and the high-limit thermostat was connected in series with the heating element. Sample s1-9 did not ignite. With an intake airflow of 4 sfpm and the high-limit thermostat connected 119

133 in series with the heating element, samples at or 1 inch from the edge of the heater housing did not ignite. 7 Lint Sample Intake High Limit, Air flow 22 High-Lmit Opens 6 s3-19 T2 s1-8 T2 s3-19 T1 Temperature (C) s3-19 T1 s1-8 T1 s3-19 s1-9 T2 s3-2 T2 s1-8 T3 s3-19 T2 s3-19 T3 s3-2 T1 s3-2 T2 s3-2 T3 s1-8 T1 s1-8 T2 s1-8 T3 s1-9 T1 1 s1-9 T2 s1-9 T Time (seconds) Figure 97. Lint Samples Tested 1 Inch from the Heater Edge with Airflow Sample s1-1 was placed 1 inch from the heater housing edge; the airflow was set at 4 sfpm and the high-limit thermostat was bypassed. The sample ignited. Figure 98 shows a comparison between samples s1-9 and s1-1. The figure shows that sample s1-1 ignited approximately 25 seconds before the high-limit thermostat operated for sample s

134 Lint Sample Intake at 1 inch with and without High Limit, Air flow 4 sfpm 22 power turned off s1-1 T2 High-Lmit Opens Temperature (C) s1-9 T1 s1-9 T2 sample 1-9 ignites s1-9 T3 s1-1 T3 s1-1 T1 s1-9 T1 s1-9 T2 s1-9 T3 s1-1 T1 s1-1 T2 s1-1 T Time (seconds) Figure 98. Samples Tested at 1 Inch, With and Without High-Limit Thermostat Lint Samples Ingested into the Heater The ignition characteristics of lint were evaluated by observing the heater exhaust as lint samples of varying weights and sizes were ingested into the heater box. The test setup was designed to allow the airflow into the heater intake to carry lint samples past the heating element and to exhaust any embers through a 4 diameter glass tube. In these tests, lint samples ranging from.4 g to.48 g were observed igniting, and embers traveled through the exhaust vent. The following observations were made (small embers are defined as particles approximately the size of a pinhead; large embers are defined as particles close to the size of a thumb tack head): 1. Sample contacted the center heating element coil and ignited. Result Small embers exited the heater box. 2. Sample contacted the edge of the heating coil and ignited. Result Small to large embers exited the heater box. In some instances, small portions of the sample that were either flaming or contained large embers broke away. 3. Sample passed between the heating element and the heater housing. Result The sample ignited and exited the heater box. 121

135 4. Sample became lodged between the two layers of heating coils, on the insulator. Result Small embers exited the heater housing, and unburned lint stayed lodged on the insulator Ignition of Target Material Downstream from the Heater Exhaust Four tests were conducted to determine whether embers that exited the heater could ignite target materials placed downstream of the heater exhaust. The target material was either 1% cotton lint or 1% cotton terry towel. In all four tests conducted, the target material ignited. The test results showed that fewer ingested lint samples were required to ignite the 1% lint target material than the 1% cotton terry target material. The target material of 1% cotton lint ignited on the fifth sample of.1 g, the sixth sample of.2 g, and the tenth sample of.1 g ingested into the heater box. The target material of 1% cotton terry ignited on the eighth sample of.3 g. When the target material ignited, the target material would produce additional embers that were carried through the exhaust vent. The 1% cotton terry towel burned substantially longer than the lint target material. The terry towel also produced significantly more embers, which were carried further into the exhaust venting High-Limit Thermostat Analysis Two high-limit thermostats failed during the course of testing. The life histories of both thermostats were compiled and are presented in Table 2: Column 1 (High Limit Thermostat) labels the thermostats (HLT1 represents the highlimit thermostat that failed first, and HLT2 represents the second). Column 2 (Life Below 1 C) lists the approximate number of hours to which the high-limit thermostat was exposed to the following conditions: The dryer operated on the primary thermostat or it operated while the temperature at T1 (intake air into the heater box) was below 1 C. This would include testing (Task 2) in which the dryer operated with an unblocked exhaust vent or with a partially-blocked exhaust vent, and it would include testing (Task 4) with just the heater box in the unblocked and partially blocked conditions. Column 3 (Life Above 1 C) lists the approximate number of hours to which the high-limit thermostat was exposed to the following conditions: The high-limit thermostat was bypassed or the temperature at T1 (intake air into the heater box) was above 1 C. This would only include testing in which the heater box was installed in the test apparatus. Column 4 (Number of Cycles) lists the number of times the high-limit thermostat cycled; one cycle represents opening and closing of the high-limit thermostat. The high-limit cycling duration was not included in either of the total times listed in column 2 and

136 Table 2. Life History of Failed High-Limit Thermostats High Limit Life Below Life Above Number of Cycles Thermostat 1 C 1 C HLT1 9.3 hours.79 hours 47 HLT hours.2 hours High-Limit Thermostat 1 (HLT1) Analysis The first high-limit thermostat failed in June 22 during Task 4 testing. This high-limit thermostat was used in Task 2: Dryer Design A Document Dryer Operating Characteristics, and Task 4: Determine Characteristics Required for Lint Ignition. The high-limit thermostat was connected in series with the heating element when it was discovered that it had failed to open. Figure 99 shows the thermocouple data for test Sample s3-13 when the high-limit first failed to operate. The figure shows the sample ignited, which was unexpected, and the T1 temperature rose to almost 2 C. 6 Sample s3-13 High Limit in Series.19g, 2" from heater box edge, ignition June Temperature (C) Time (seconds) T3 Sample Ignites T2 T T Air Flow (sfpm) T1 Heater Intake T2 Heater Exhaust T3 Liint Sample T4 Exhaust Vent Heater T5 Exhaust Vent Outside Heater T6 Chamber T7 Room Ambient V1 Airflow into Heater V2 Airflow External Heater Figure 99. Sample s3-13 Test with the High Limit Thermostat in Series Figure 1 shows a comparison for thermocouples T1, T2, and T3 between the failed high-limit thermostat (HLT1) and a replacement high-limit thermostat (HLT2). The 123

137 thermocouples were placed at the heater intake, the heater exhaust, and 2 inches from the heater edge for T1, T2 and T3, respectively. The figure shows T1 for HLT2 rose to approximately 15 C before the high-limit thermostat opened and disconnected power to the heating element. The temperatures for HLT1 continued to rise until power was manually turned off at approximately 17 seconds. High Limit Thermostat 1 (HLT1) and Replacement High Limit Thermostat 2 (HLT2) June 22 6 T2 HLT1 Heater Exhaust Power to the heater manually turned off 5 Temperature (C) High-Limit Opens T1 HLT2 Inatke Heater T3 HLT2 2" from Heater Intake T2 HL2 Heater Exhaust Time (seconds) T1 HLT1 Intake Heater T3 HLT1 2" from Heater Intake T1 HLT1 T2 HLT1 T3 HLT1 T1 HLT2 T2 HLT2 T3 HLT Figure 1. Failed Thermostat compared to a Replacement Thermostat The high-limit thermostat was removed and x-rayed as shown in Figure 11. The highlimit thermostat uses a bi-metal disc that pops at a certain set temperature. When the concave disc pops and becomes convex, it pushes on a rod that sequentially opens the contacts. The x-rays did not reveal any abnormalities. 124

138 Rod Contacts Bi-Metal Disc Figure 11. X-ray Images of High-Limit Thermostat 1 (HLT1) The high-limit thermostat was opened and examined under a microscope, as shown in Figure 12. It appeared that the contacts were welded together, but this could not be confirmed. Moving Contact Fixed Contact Fixed Contact Moving Contact Figure 12. Contacts for High-Limit Thermostat 1 (HLT1) Figure 13 shows the surface of the contact pads. Pitting and scorch marks are visible and indicate there was a concentration of high heat. Figures 12 and 13 also show that approximately 5% to 6% of the contact pad surfaces were making contact when the thermostat was in the closed position. 125

139 Fixed Contact Moving Contact Figure 13. Contact Pads for High-Limit Thermostat 1 (HLT1) 126

140 High-Limit Thermostat 2 (HLT2) Analysis The second high-limit thermostat failed during tests in July 22 during Task 4: Determine Characteristics Required for Lint Ignition. This high-limit thermostat was used in Task 3 Monitor Lint Distribution testing and as a replacement thermostat for Task 4: Characteristics for Lint Ignition tests. The high-limit thermostat was in series with the heating element when it was discovered that it had failed to open. Figure 14 shows the thermocouple data for test Sample s5-14 when the high-limit failed to operate. The figure shows the T2 temperature rose to almost 6 C. (The T1 thermocouple and V1 anemometer were not in the heater intake air stream during this test.) s5-14 July 12, T2 Heater Exhaust Power to the heater manually turned off Temperature (C) T3 Lint Sample T4 Heater Exhaust Airflow (sfpm) Time (seconds) T1 Heater Intake T2 Heater Exhaust T3 Liint Sample T4 Exhaust Vent Heater T5 Exhaust Vent Outside Heater T6 Chamber T7 Room Ambient V1 Airflow into Heater V2 Airflow External Heater Figure 14. Sample s5-14 with High Limit Thermostat in Series Figure 15 shows a comparison of temperature data from thermocouples T2 and T3 for the high-limit thermostat before it failed (HLT2 Before) and after it failed (HLT2 After). The thermocouples were placed at the heater exhaust and 2 inches from the heater edge for T2 and T3, respectively. (The T1 thermocouple and V1 anemometer were not in the heater intake air stream during the HL2 After test.) The figure shows T1 temperature for HL2 Before rose to approximately 15 C before the high-limit thermostat opened and disconnected power to the heating element. The sample lint in the HLT2 After test ignited unexpectedly approximately 55 seconds after powering the heater. 127

141 The temperatures for HLT2 After continued to rise until power was manually turned off at approximately 17 seconds. High Limit Thermostat 2 Before Failure (HLT2 Before) and After Failure (HLT2 After) 6 5 T3 HLT2 (after) 2" from Heater Intake Power to the heater manually turned off Temperature (C) High-Limit Opens (before) Sample Ignites T3 HLT2 (before) 2" from Heater Intake T1 HLT2 after Time (seconds) T2 HLT2 (after) Heater Exhaust T1 HLT2 (before) Intake Heater T2 HLT2 (before) Heater Exhaust T1 HLT1 after T2 HLT1 after T3 HLT1 after T1 HLT2 before T2 HLT2 before T3 HLT2 before Figure 15. High-Limit Thermostat 2 (HLT2) before and after Failure The high-limit thermostat was removed and x-rayed, as shown in Figure 16. The theory of operation for the second failed thermostat, HLT2, is the same as that for the first failed thermostat, HLT1. The x-rays did not reveal any abnormalities. 128

142 Rod Contacts Bi-Metal Disc Figure 16. X-ray Images of High-Limit Thermostat 2 (HLT2) A portion of the thermostat housing was removed to reveal the contacts and the rod, as shown in Figure 17. From visual examination, it could not be determined if the contacts were welded together or not. This sample was not disassembled, to maintain the operational integrity of the thermostat; and an inspection of the contact pad surfaces was not performed. Rod Moving Contact Fixed Contact Fixed Contact Moving Contact Figure 17. Internal View for High-Limit Thermostat 2 Before conducting tests of the failed thermostat HLT2, tests were first conducted on a new high limit thermostat of the same make and model. The setup included a heat gun (heat source), a microscope, and dual channel logging thermocouple temperature meter, as shown in Figure 18. Thermocouple 1 was placed at the face of the enclosure, near the bi-metal disc. Thermocouple 2 was used as an indicator of when the contacts opened and closed. 129

143 Microscope T1 Thermocouple T2 Thermocouple High-Limit Thermostat Heat Source Figure 18. Setup for Testing High-Limit Thermostats A small grinder was used to produce a small hole to reveal the internal conponents around the contacts of the new thermostat. Figure 19 shows a before picture (contacts in the normally-closed position) and an after picture (contacts opened). The after picture shows the rod had pushed on the arm and caused the contacts to separate. Bi-Metal Disc Normal Rod Bi-Metal Disc Flipped Contacts are Separated (opened) Figure 19. New High Limit Thermostat Figure 11 shows the temperature data for T1 thermocouple at the bi-metal disc and T2 at the contacts for the new thermostat. As mentioned previously, T2 was used as an idicator of when the contacts opened and closed. The temperature values for T2 were translated to and 13

144 1, where indicates a closed contact and 1 indicates an open contact. The contacts opened when T1 was at C and closed when T1 was at 49.6 C. New High Limit Thermostat Test New Temperature (C) Contacts ( closed, 1 open) Time (seconds) T1 Contacts Figure 11. Thermocouple Data for the New High-Limit Thermostat The second failed high limit thermostat (HLT2) was tested in the setup. An audible click could be heard when the bi-metal disc popped. When the bi-metal disc popped, it pushed on the rod, which should have separated the contacts. However, the T2 signal did not indicate a separation of the contacts. The resistance was measured at. to.1 ohms throughout the test. Figure 111 shows a picture before the bi-metal disc popped and an after picture. The before picture (Bi-Metal Disc Normal ) shows a small gap between the rod and the arm for the moving contact. The after figure shows the rod had moved and contacted the arm which should separate the contacts. However, the contacts remained closed. Bi-Metal Disc Normal Rod Bi-Metal Disc Flipped Contacts remain Closed Figure 111. Second Failed High Limit Thermostat, HLT2 131